AlexHung29629/openimages_objdet · Datasets at Hugging Face

messages listlengths 3 3
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03f6288f49430b0c.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Knife</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.16</ymin>\n <xmax>0.49</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/018e2d22c8b97099.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0.76</ymin>\n <xmax>0.26</xmax>\n <ymax>0.87</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.77</ymin>\n <xmax>0.44</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.41</xmin>\n <ymin>0.76</ymin>\n <xmax>0.53</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0054c91397f2fe05.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Mobile phone</name>\n <bndbox>\n <xmin>0.21</xmin>\n <ymin>0.27</ymin>\n <xmax>0.78</xmax>\n <ymax>0.6</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0454f5b879250fdd.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Suit</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.21</ymin>\n <xmax>0.4</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08e407dd23779aa9.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tire</name>\n <bndbox>\n <xmin>0.33</xmin>\n <ymin>0.71</ymin>\n <xmax>0.39</xmax>\n <ymax>0.76</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/000468989006adbd.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.39</xmin>\n <ymin>0.42</ymin>\n <xmax>0.49</xmax>\n <ymax>0.47</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.57</xmin>\n <ymin>0.4</ymin>\n <xmax>0.68</xmax>\n <ymax>0.44</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.18</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sun hat</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.19</ymin>\n <xmax>0.82</xmax>\n <ymax>0.44</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.36</xmin>\n <ymin>0.31</ymin>\n <xmax>0.77</xmax>\n <ymax>0.61</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human nose</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.42</ymin>\n <xmax>0.58</xmax>\n <ymax>0.51</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hand</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0.74</ymin>\n <xmax>0.49</xmax>\n <ymax>0.94</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01cfb3c9537cdff9.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Food</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.47</ymin>\n <xmax>0.83</xmax>\n <ymax>0.74</ymax>\n </bndbox>\n </object>\n <object>\n <name>Food</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.33</ymin>\n <xmax>0.53</xmax>\n <ymax>0.54</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fork</name>\n <bndbox>\n <xmin>0.78</xmin>\n <ymin>0.58</ymin>\n <xmax>1</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06cf9bd412ce188c.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Jeans</name>\n <bndbox>\n <xmin>0.81</xmin>\n <ymin>0.63</ymin>\n <xmax>0.89</xmax>\n <ymax>0.8</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.79</xmin>\n <ymin>0.77</ymin>\n <xmax>0.84</xmax>\n <ymax>0.81</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01cde177c49a0d15.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Saucer</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.85</ymin>\n <xmax>0.28</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Kettle</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.54</ymin>\n <xmax>0.51</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.01</ymin>\n <xmax>0.35</xmax>\n <ymax>0.23</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.88</xmin>\n <ymin>0.09</ymin>\n <xmax>1</xmax>\n <ymax>0.29</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08859fd788da1ae3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Hamburger</name>\n <bndbox>\n <xmin>0.16</xmin>\n <ymin>0.23</ymin>\n <xmax>0.74</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Baked goods</name>\n <bndbox>\n <xmin>0.16</xmin>\n <ymin>0.27</ymin>\n <xmax>0.72</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07e45d793f61a648.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.74</xmin>\n <ymin>0.38</ymin>\n <xmax>0.91</xmax>\n <ymax>0.63</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00737efc6f6a967e.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.51</ymin>\n <xmax>0.22</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.51</ymin>\n <xmax>0.38</xmax>\n <ymax>0.76</ymax>\n </bndbox>\n </object>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.4</xmin>\n <ymin>0.52</ymin>\n <xmax>0.52</xmax>\n <ymax>0.74</ymax>\n </bndbox>\n </object>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.51</ymin>\n <xmax>0.77</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.79</xmin>\n <ymin>0.51</ymin>\n <xmax>0.87</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.89</xmin>\n <ymin>0.52</ymin>\n <xmax>0.96</xmax>\n <ymax>0.68</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/04d0facfa46d701f.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.17</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02a0052b6f0fc8fd.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Paddle</name>\n <bndbox>\n <xmin>0.19</xmin>\n <ymin>0.56</ymin>\n <xmax>0.41</xmax>\n <ymax>0.67</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02fcc89f4e1269c2.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Fashion accessory</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.03</ymin>\n <xmax>0.66</xmax>\n <ymax>0.07</ymax>\n </bndbox>\n </object>\n <object>\n <name>Girl</name>\n <bndbox>\n <xmin>0.34</xmin>\n <ymin>0</ymin>\n <xmax>0.88</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06ad4e51ed358293.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Train</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/002e55ad1b03b9f0.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.1</ymin>\n <xmax>0.86</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human nose</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.23</ymin>\n <xmax>0.73</xmax>\n <ymax>0.35</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/050fa6f013133f7a.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Broccoli</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.11</ymin>\n <xmax>0.57</xmax>\n <ymax>0.45</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01bd8071242b5d59.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Camera</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07b0d7fda7e2eea8.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.79</ymin>\n <xmax>0.49</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.52</xmin>\n <ymin>0.79</ymin>\n <xmax>0.59</xmax>\n <ymax>0.83</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00dc48131468b9f0.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Woman</name>\n <bndbox>\n <xmin>0.17</xmin>\n <ymin>0.12</ymin>\n <xmax>0.39</xmax>\n <ymax>0.46</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.55</ymin>\n <xmax>0.09</xmax>\n <ymax>0.64</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.14</ymin>\n <xmax>0.31</xmax>\n <ymax>0.18</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/072e68abf4f016ec.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bed</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.7</ymin>\n <xmax>0.96</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n <object>\n <name>Cat</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.35</ymin>\n <xmax>0.62</xmax>\n <ymax>0.8</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03981a436416648c.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cowboy hat</name>\n <bndbox>\n <xmin>0.13</xmin>\n <ymin>0.09</ymin>\n <xmax>0.75</xmax>\n <ymax>0.38</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fedora</name>\n <bndbox>\n <xmin>0.09</xmin>\n <ymin>0.08</ymin>\n <xmax>0.76</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sun hat</name>\n <bndbox>\n <xmin>0.08</xmin>\n <ymin>0.08</ymin>\n <xmax>0.77</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0001e27f4b156f49.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.32</ymin>\n <xmax>0.47</xmax>\n <ymax>0.34</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.52</xmin>\n <ymin>0.31</ymin>\n <xmax>0.54</xmax>\n <ymax>0.33</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.31</ymin>\n <xmax>0.56</xmax>\n <ymax>0.32</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.69</xmin>\n <ymin>0.57</ymin>\n <xmax>0.69</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.7</xmin>\n <ymin>0.6</ymin>\n <xmax>0.72</xmax>\n <ymax>0.61</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human mouth</name>\n <bndbox>\n <xmin>0.47</xmin>\n <ymin>0.38</ymin>\n <xmax>0.48</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human mouth</name>\n <bndbox>\n <xmin>0.54</xmin>\n <ymin>0.36</ymin>\n <xmax>0.56</xmax>\n <ymax>0.37</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human mouth</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.63</ymin>\n <xmax>0.69</xmax>\n <ymax>0.66</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human body</name>\n <bndbox>\n <xmin>0.72</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Mammal</name>\n <bndbox>\n <xmin>0.33</xmin>\n <ymin>0.22</ymin>\n <xmax>0.51</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Mammal</name>\n <bndbox>\n <xmin>0.5</xmin>\n <ymin>0.07</ymin>\n <xmax>0.54</xmax>\n <ymax>0.15</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03a23823ea019ba6.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Table</name>\n <bndbox>\n <xmin>0.75</xmin>\n <ymin>0.76</ymin>\n <xmax>0.99</xmax>\n <ymax>0.97</ymax>\n </bndbox>\n </object>\n <object>\n <name>Chair</name>\n <bndbox>\n <xmin>0.44</xmin>\n <ymin>0.76</ymin>\n <xmax>0.57</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n <object>\n <name>Wheel</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.9</ymin>\n <xmax>0.16</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.17</xmin>\n <ymin>0.22</ymin>\n <xmax>0.33</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.34</xmin>\n <ymin>0.22</ymin>\n <xmax>0.46</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.64</xmin>\n <ymin>0.23</ymin>\n <xmax>0.79</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06f5e60505dec0c6.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Box</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0.1</ymin>\n <xmax>0.81</xmax>\n <ymax>0.86</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/077f5162b0fe17ec.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Furniture</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08954de74350e0a6.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.06</ymin>\n <xmax>0.49</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07c5c0c1b9f9cda1.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cosmetics</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.54</ymin>\n <xmax>0.55</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00078712936e5c4b.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Butterfly</name>\n <bndbox>\n <xmin>0.11</xmin>\n <ymin>0.08</ymin>\n <xmax>0.73</xmax>\n <ymax>0.33</ymax>\n </bndbox>\n </object>\n <object>\n <name>Butterfly</name>\n <bndbox>\n <xmin>0.16</xmin>\n <ymin>0.35</ymin>\n <xmax>0.71</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n <object>\n <name>Butterfly</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0.61</ymin>\n <xmax>0.68</xmax>\n <ymax>0.82</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02f322db7a5996b5.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cat</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.41</ymin>\n <xmax>0.56</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/015155a62fd65a13.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Pumpkin</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.49</ymin>\n <xmax>0.83</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n <object>\n <name>Pumpkin</name>\n <bndbox>\n <xmin>0.11</xmin>\n <ymin>0.05</ymin>\n <xmax>0.49</xmax>\n <ymax>0.54</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/036b8e5b2408f356.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.82</xmin>\n <ymin>0.79</ymin>\n <xmax>0.94</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0014c3b48667efde.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Toy</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.57</ymin>\n <xmax>0.39</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n <object>\n <name>Girl</name>\n <bndbox>\n <xmin>0.33</xmin>\n <ymin>0.1</ymin>\n <xmax>0.74</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/026ace2555b5f02d.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Spoon</name>\n <bndbox>\n <xmin>0.47</xmin>\n <ymin>0.17</ymin>\n <xmax>0.65</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n <object>\n <name>Spoon</name>\n <bndbox>\n <xmin>0.64</xmin>\n <ymin>0.17</ymin>\n <xmax>0.9</xmax>\n <ymax>0.6</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07c4c14d0b894f77.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.04</xmin>\n <ymin>0</ymin>\n <xmax>0.68</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bottle</name>\n <bndbox>\n <xmin>0.7</xmin>\n <ymin>0.79</ymin>\n <xmax>0.8</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0285f617f6cdf4ba.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.28</xmin>\n <ymin>0.29</ymin>\n <xmax>0.76</xmax>\n <ymax>0.82</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.09</ymin>\n <xmax>0.61</xmax>\n <ymax>0.24</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00421577770d9db5.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Drink</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.61</ymin>\n <xmax>0.97</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/022f06ac4b731813.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Crown</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.07</ymin>\n <xmax>0.82</xmax>\n <ymax>0.27</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/051233a32a0bbf2e.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0316ff2baf27cf29.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Giraffe</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.15</ymin>\n <xmax>0.6</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0692df6384563bd0.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.16</xmin>\n <ymin>0.22</ymin>\n <xmax>0.31</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/045707a371224f49.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.57</ymin>\n <xmax>0.36</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>Building</name>\n <bndbox>\n <xmin>0.11</xmin>\n <ymin>0.09</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.69</ymin>\n <xmax>0.02</xmax>\n <ymax>0.76</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.55</ymin>\n <xmax>0.03</xmax>\n <ymax>0.6</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.85</ymin>\n <xmax>0.05</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.03</xmin>\n <ymin>0.68</ymin>\n <xmax>0.07</xmax>\n <ymax>0.75</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.53</ymin>\n <xmax>0.08</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.07</xmin>\n <ymin>0.84</ymin>\n <xmax>0.11</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.08</xmin>\n <ymin>0.65</ymin>\n <xmax>0.12</xmax>\n <ymax>0.73</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.1</xmin>\n <ymin>0.5</ymin>\n <xmax>0.14</xmax>\n <ymax>0.57</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.63</ymin>\n <xmax>0.22</xmax>\n <ymax>0.86</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.18</xmin>\n <ymin>0.37</ymin>\n <xmax>0.23</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0.6</ymin>\n <xmax>0.27</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0.33</ymin>\n <xmax>0.31</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.34</xmin>\n <ymin>0.28</ymin>\n <xmax>0.41</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.56</ymin>\n <xmax>0.4</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.29</ymin>\n <xmax>0.66</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.56</ymin>\n <xmax>0.65</xmax>\n <ymax>0.83</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.35</ymin>\n <xmax>0.72</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.6</ymin>\n <xmax>0.73</xmax>\n <ymax>0.86</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.74</xmin>\n <ymin>0.4</ymin>\n <xmax>0.77</xmax>\n <ymax>0.57</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.74</xmin>\n <ymin>0.65</ymin>\n <xmax>0.79</xmax>\n <ymax>0.87</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.8</xmin>\n <ymin>0.51</ymin>\n <xmax>0.82</xmax>\n <ymax>0.63</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.81</xmin>\n <ymin>0.68</ymin>\n <xmax>0.84</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.85</xmin>\n <ymin>0.54</ymin>\n <xmax>0.87</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/030963627750d750.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.07</ymin>\n <xmax>0.75</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0112a8b3ef79fa20.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Golf ball</name>\n <bndbox>\n <xmin>0.44</xmin>\n <ymin>0.59</ymin>\n <xmax>0.46</xmax>\n <ymax>0.62</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/023b74dc25cdbbe2.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Flower</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01047e9822575c85.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Pastry</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.41</ymin>\n <xmax>0.53</xmax>\n <ymax>0.65</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00c25ea2a49b6908.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.89</ymin>\n <xmax>0.4</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Building</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00d3d507e389f463.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bee</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.48</ymin>\n <xmax>0.7</xmax>\n <ymax>0.6</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/04e603dbe6e114fa.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.08</xmin>\n <ymin>0.25</ymin>\n <xmax>0.17</xmax>\n <ymax>0.64</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/044f24685a8f2a22.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Moths and butterflies</name>\n <bndbox>\n <xmin>0.17</xmin>\n <ymin>0.12</ymin>\n <xmax>0.83</xmax>\n <ymax>0.82</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07cb1062d1cc55f3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Jeans</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.71</ymin>\n <xmax>0.67</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.17</ymin>\n <xmax>0.7</xmax>\n <ymax>0.41</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.92</ymin>\n <xmax>0.62</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.62</xmin>\n <ymin>0.94</ymin>\n <xmax>0.64</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02f9204ee8ad6507.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Boy</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.44</ymin>\n <xmax>0.61</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.58</ymin>\n <xmax>0.45</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.44</xmin>\n <ymin>0.59</ymin>\n <xmax>0.62</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.4</xmin>\n <ymin>0.47</ymin>\n <xmax>0.46</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07b4cd46a385b4f8.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tart</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0296272c25074ec1.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Vehicle registration plate</name>\n <bndbox>\n <xmin>0.72</xmin>\n <ymin>0.37</ymin>\n <xmax>0.81</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0.12</ymin>\n <xmax>0.33</xmax>\n <ymax>0.3</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.12</ymin>\n <xmax>0.47</xmax>\n <ymax>0.31</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.89</xmin>\n <ymin>0.1</ymin>\n <xmax>1</xmax>\n <ymax>0.23</ymax>\n </bndbox>\n </object>\n <object>\n <name>Car</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0.05</ymin>\n <xmax>0.92</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0539739ee5ce9be0.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Woman</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.54</ymin>\n <xmax>1</xmax>\n <ymax>0.97</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0418b55b2b74d429.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Suit</name>\n <bndbox>\n <xmin>0.01</xmin>\n <ymin>0.29</ymin>\n <xmax>0.2</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.03</xmin>\n <ymin>0.29</ymin>\n <xmax>0.11</xmax>\n <ymax>0.44</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.41</xmin>\n <ymin>0.34</ymin>\n <xmax>0.51</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.71</xmin>\n <ymin>0.33</ymin>\n <xmax>0.79</xmax>\n <ymax>0.5</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0296768e960a1e8a.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.09</ymin>\n <xmax>1</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03548fd47986828d.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.75</xmin>\n <ymin>0.76</ymin>\n <xmax>0.91</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>House</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.38</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.78</ymin>\n <xmax>0.38</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.67</ymin>\n <xmax>0.39</xmax>\n <ymax>0.77</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.39</xmin>\n <ymin>0.58</ymin>\n <xmax>0.45</xmax>\n <ymax>0.63</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.67</ymin>\n <xmax>0.53</xmax>\n <ymax>0.77</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.78</ymin>\n <xmax>0.54</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.61</ymin>\n <xmax>0.68</xmax>\n <ymax>0.68</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.75</ymin>\n <xmax>0.68</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.71</xmin>\n <ymin>0.61</ymin>\n <xmax>0.9</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06269adab3d0dbfa.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.07</ymin>\n <xmax>0.44</xmax>\n <ymax>0.55</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.3</xmax>\n <ymax>0.74</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.3</ymin>\n <xmax>0.33</xmax>\n <ymax>0.57</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.33</xmin>\n <ymin>0.43</ymin>\n <xmax>0.37</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.44</xmin>\n <ymin>0.5</ymin>\n <xmax>0.47</xmax>\n <ymax>0.58</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.5</xmin>\n <ymin>0.5</ymin>\n <xmax>0.54</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.53</xmin>\n <ymin>0.44</ymin>\n <xmax>0.56</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.48</ymin>\n <xmax>0.62</xmax>\n <ymax>0.58</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.57</xmin>\n <ymin>0.32</ymin>\n <xmax>0.66</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.41</ymin>\n <xmax>0.69</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.73</xmin>\n <ymin>0.37</ymin>\n <xmax>0.81</xmax>\n <ymax>0.5</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.77</xmin>\n <ymin>0.45</ymin>\n <xmax>0.81</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.81</xmin>\n <ymin>0.45</ymin>\n <xmax>0.86</xmax>\n <ymax>0.55</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.86</xmin>\n <ymin>0.44</ymin>\n <xmax>0.9</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.9</xmin>\n <ymin>0.33</ymin>\n <xmax>1</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.07</ymin>\n <xmax>0.44</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02553b1d3984ebff.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Pig</name>\n <bndbox>\n <xmin>0.17</xmin>\n <ymin>0.15</ymin>\n <xmax>0.74</xmax>\n <ymax>0.82</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/001ce300279f1a81.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Door</name>\n <bndbox>\n <xmin>0.9</xmin>\n <ymin>0.27</ymin>\n <xmax>1</xmax>\n <ymax>0.87</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/010db899406b65a6.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.01</xmin>\n <ymin>0.53</ymin>\n <xmax>0.03</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.64</ymin>\n <xmax>0.05</xmax>\n <ymax>0.67</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.43</ymin>\n <xmax>0.08</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.65</ymin>\n <xmax>0.08</xmax>\n <ymax>0.68</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.45</ymin>\n <xmax>0.17</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.66</ymin>\n <xmax>0.44</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.68</ymin>\n <xmax>0.47</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.52</xmin>\n <ymin>0.68</ymin>\n <xmax>0.55</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.47</ymin>\n <xmax>0.6</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.68</ymin>\n <xmax>0.61</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.87</xmin>\n <ymin>0.46</ymin>\n <xmax>0.9</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.89</xmin>\n <ymin>0.72</ymin>\n <xmax>0.91</xmax>\n <ymax>0.74</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.93</xmin>\n <ymin>0.68</ymin>\n <xmax>0.94</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0274165b127bd47e.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>House</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.31</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>House</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.04</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/044b37bca139f280.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Vehicle</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.52</ymin>\n <xmax>0.07</xmax>\n <ymax>0.6</ymax>\n </bndbox>\n </object>\n <object>\n <name>Vehicle</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0.27</ymin>\n <xmax>0.69</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02fe016b54c838f3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Fedora</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.18</ymin>\n <xmax>0.73</xmax>\n <ymax>0.38</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sun hat</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.18</ymin>\n <xmax>0.73</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/020294a97cbd1bba.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bottle</name>\n <bndbox>\n <xmin>0.04</xmin>\n <ymin>0.04</ymin>\n <xmax>0.26</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bottle</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.13</ymin>\n <xmax>0.52</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bottle</name>\n <bndbox>\n <xmin>0.54</xmin>\n <ymin>0.27</ymin>\n <xmax>0.74</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bottle</name>\n <bndbox>\n <xmin>0.76</xmin>\n <ymin>0.4</ymin>\n <xmax>0.93</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/048e97a148df0741.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.67</ymin>\n <xmax>0.1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.09</xmin>\n <ymin>0.7</ymin>\n <xmax>0.32</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.77</ymin>\n <xmax>0.51</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.57</xmin>\n <ymin>0.78</ymin>\n <xmax>0.79</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.97</xmin>\n <ymin>0.56</ymin>\n <xmax>1</xmax>\n <ymax>0.76</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/062d2386effbc304.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Wheel</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02a940159cd39532.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.44</xmin>\n <ymin>0.32</ymin>\n <xmax>0.6</xmax>\n <ymax>0.37</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0</ymin>\n <xmax>0.9</xmax>\n <ymax>0.37</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human head</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0.01</ymin>\n <xmax>0.91</xmax>\n <ymax>0.65</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.3</ymin>\n <xmax>0.37</xmax>\n <ymax>0.35</ymax>\n </bndbox>\n </object>\n <object>\n <name>Shirt</name>\n <bndbox>\n <xmin>0.19</xmin>\n <ymin>0.47</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human mouth</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.46</ymin>\n <xmax>0.55</xmax>\n <ymax>0.55</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human ear</name>\n <bndbox>\n <xmin>0.78</xmin>\n <ymin>0.28</ymin>\n <xmax>0.92</xmax>\n <ymax>0.44</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.1</ymin>\n <xmax>0.79</xmax>\n <ymax>0.64</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human nose</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.33</ymin>\n <xmax>0.48</xmax>\n <ymax>0.45</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03a6437d5e138df3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Jeans</name>\n <bndbox>\n <xmin>0.48</xmin>\n <ymin>0.01</ymin>\n <xmax>0.78</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01cfd918cf92072d.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.51</xmin>\n <ymin>0.47</ymin>\n <xmax>0.58</xmax>\n <ymax>0.61</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06f11bc57307517a.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.73</ymin>\n <xmax>0.1</xmax>\n <ymax>0.85</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.68</ymin>\n <xmax>0.31</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.7</ymin>\n <xmax>0.34</xmax>\n <ymax>0.8</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0.7</ymin>\n <xmax>0.41</xmax>\n <ymax>0.78</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.41</xmin>\n <ymin>0.77</ymin>\n <xmax>0.47</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.77</ymin>\n <xmax>0.68</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.77</ymin>\n <xmax>0.68</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.72</xmin>\n <ymin>0.79</ymin>\n <xmax>0.8</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.86</xmin>\n <ymin>0.8</ymin>\n <xmax>0.94</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0815654513ba23b3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Car</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.55</ymin>\n <xmax>0.2</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n <object>\n <name>Car</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.57</ymin>\n <xmax>0.42</xmax>\n <ymax>0.68</ymax>\n </bndbox>\n </object>\n <object>\n <name>Car</name>\n <bndbox>\n <xmin>0.77</xmin>\n <ymin>0.56</ymin>\n <xmax>1</xmax>\n <ymax>0.86</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0145aabe9dce2c5a.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.23</xmin>\n <ymin>0.48</ymin>\n <xmax>0.3</xmax>\n <ymax>0.73</ymax>\n </bndbox>\n </object>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.46</ymin>\n <xmax>0.39</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.82</xmin>\n <ymin>0.57</ymin>\n <xmax>0.92</xmax>\n <ymax>0.73</ymax>\n </bndbox>\n </object>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.92</xmin>\n <ymin>0.61</ymin>\n <xmax>1</xmax>\n <ymax>0.73</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00d9a083e1cba929.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Screwdriver</name>\n <bndbox>\n <xmin>0.27</xmin>\n <ymin>0.45</ymin>\n <xmax>0.3</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0042d5340087fb4d.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Girl</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.42</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/038cf4c2eb7ea83b.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Wine glass</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0</ymin>\n <xmax>0.7</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Wine glass</name>\n <bndbox>\n <xmin>0.8</xmin>\n <ymin>0.34</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/038596110c8e6992.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.64</ymin>\n <xmax>0.47</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Table</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.56</ymin>\n <xmax>0.61</xmax>\n <ymax>0.8</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.09</xmin>\n <ymin>0.12</ymin>\n <xmax>0.14</xmax>\n <ymax>0.24</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.18</xmin>\n <ymin>0.31</ymin>\n <xmax>0.25</xmax>\n <ymax>0.38</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.5</ymin>\n <xmax>0.04</xmax>\n <ymax>0.67</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.1</xmin>\n <ymin>0.1</ymin>\n <xmax>0.13</xmax>\n <ymax>0.14</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0.43</ymin>\n <xmax>0.24</xmax>\n <ymax>0.51</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.75</xmin>\n <ymin>0.26</ymin>\n <xmax>0.77</xmax>\n <ymax>0.29</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.88</xmin>\n <ymin>0.27</ymin>\n <xmax>0.9</xmax>\n <ymax>0.31</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06a55a5c51371377.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.43</ymin>\n <xmax>0.7</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00915674b39ba67c.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.08</xmin>\n <ymin>0</ymin>\n <xmax>0.64</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.87</xmin>\n <ymin>0.09</ymin>\n <xmax>1</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/04e96e5104d98896.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Woman</name>\n <bndbox>\n <xmin>0.01</xmin>\n <ymin>0.41</ymin>\n <xmax>0.39</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.44</ymin>\n <xmax>0.19</xmax>\n <ymax>0.63</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/068bf4f046cc95e6.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.07</ymin>\n <xmax>0.48</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.15</ymin>\n <xmax>0.63</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/026fd9ac9f553bd3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cattle</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0.17</ymin>\n <xmax>0.66</xmax>\n <ymax>0.63</ymax>\n </bndbox>\n </object>\n <object>\n <name>Cattle</name>\n <bndbox>\n <xmin>0.82</xmin>\n <ymin>0.5</ymin>\n <xmax>1</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0458e49dbe451311.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Dress</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.42</ymin>\n <xmax>0.59</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tiara</name>\n <bndbox>\n <xmin>0.4</xmin>\n <ymin>0.21</ymin>\n <xmax>0.47</xmax>\n <ymax>0.28</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08cc19b1c57067bd.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.04</ymin>\n <xmax>0.92</xmax>\n <ymax>0.57</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]
[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are fully visible — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be normalized to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output only the results in the following XML format, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/041b524627acae82.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tiger</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.04</ymin>\n <xmax>0.99</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03f6288f49430b0c.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Knife</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.16</ymin>\n <xmax>0.49</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/018e2d22c8b97099.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0.76</ymin>\n <xmax>0.26</xmax>\n <ymax>0.87</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.77</ymin>\n <xmax>0.44</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.41</xmin>\n <ymin>0.76</ymin>\n <xmax>0.53</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0054c91397f2fe05.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Mobile phone</name>\n <bndbox>\n <xmin>0.21</xmin>\n <ymin>0.27</ymin>\n <xmax>0.78</xmax>\n <ymax>0.6</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0454f5b879250fdd.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Suit</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.21</ymin>\n <xmax>0.4</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08e407dd23779aa9.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tire</name>\n <bndbox>\n <xmin>0.33</xmin>\n <ymin>0.71</ymin>\n <xmax>0.39</xmax>\n <ymax>0.76</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/000468989006adbd.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.39</xmin>\n <ymin>0.42</ymin>\n <xmax>0.49</xmax>\n <ymax>0.47</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.57</xmin>\n <ymin>0.4</ymin>\n <xmax>0.68</xmax>\n <ymax>0.44</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.18</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sun hat</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.19</ymin>\n <xmax>0.82</xmax>\n <ymax>0.44</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.36</xmin>\n <ymin>0.31</ymin>\n <xmax>0.77</xmax>\n <ymax>0.61</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human nose</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.42</ymin>\n <xmax>0.58</xmax>\n <ymax>0.51</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hand</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0.74</ymin>\n <xmax>0.49</xmax>\n <ymax>0.94</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01cfb3c9537cdff9.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Food</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.47</ymin>\n <xmax>0.83</xmax>\n <ymax>0.74</ymax>\n </bndbox>\n </object>\n <object>\n <name>Food</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.33</ymin>\n <xmax>0.53</xmax>\n <ymax>0.54</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fork</name>\n <bndbox>\n <xmin>0.78</xmin>\n <ymin>0.58</ymin>\n <xmax>1</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06cf9bd412ce188c.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Jeans</name>\n <bndbox>\n <xmin>0.81</xmin>\n <ymin>0.63</ymin>\n <xmax>0.89</xmax>\n <ymax>0.8</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.79</xmin>\n <ymin>0.77</ymin>\n <xmax>0.84</xmax>\n <ymax>0.81</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01cde177c49a0d15.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Saucer</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.85</ymin>\n <xmax>0.28</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Kettle</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.54</ymin>\n <xmax>0.51</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.01</ymin>\n <xmax>0.35</xmax>\n <ymax>0.23</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.88</xmin>\n <ymin>0.09</ymin>\n <xmax>1</xmax>\n <ymax>0.29</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08859fd788da1ae3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Hamburger</name>\n <bndbox>\n <xmin>0.16</xmin>\n <ymin>0.23</ymin>\n <xmax>0.74</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Baked goods</name>\n <bndbox>\n <xmin>0.16</xmin>\n <ymin>0.27</ymin>\n <xmax>0.72</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07e45d793f61a648.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.74</xmin>\n <ymin>0.38</ymin>\n <xmax>0.91</xmax>\n <ymax>0.63</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00737efc6f6a967e.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.51</ymin>\n <xmax>0.22</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.51</ymin>\n <xmax>0.38</xmax>\n <ymax>0.76</ymax>\n </bndbox>\n </object>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.4</xmin>\n <ymin>0.52</ymin>\n <xmax>0.52</xmax>\n <ymax>0.74</ymax>\n </bndbox>\n </object>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.51</ymin>\n <xmax>0.77</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.79</xmin>\n <ymin>0.51</ymin>\n <xmax>0.87</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0.89</xmin>\n <ymin>0.52</ymin>\n <xmax>0.96</xmax>\n <ymax>0.68</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/04d0facfa46d701f.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.17</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02a0052b6f0fc8fd.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Paddle</name>\n <bndbox>\n <xmin>0.19</xmin>\n <ymin>0.56</ymin>\n <xmax>0.41</xmax>\n <ymax>0.67</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02fcc89f4e1269c2.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Fashion accessory</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.03</ymin>\n <xmax>0.66</xmax>\n <ymax>0.07</ymax>\n </bndbox>\n </object>\n <object>\n <name>Girl</name>\n <bndbox>\n <xmin>0.34</xmin>\n <ymin>0</ymin>\n <xmax>0.88</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06ad4e51ed358293.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Train</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/002e55ad1b03b9f0.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.1</ymin>\n <xmax>0.86</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human nose</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.23</ymin>\n <xmax>0.73</xmax>\n <ymax>0.35</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/050fa6f013133f7a.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Broccoli</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.11</ymin>\n <xmax>0.57</xmax>\n <ymax>0.45</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01bd8071242b5d59.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Camera</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07b0d7fda7e2eea8.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.79</ymin>\n <xmax>0.49</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.52</xmin>\n <ymin>0.79</ymin>\n <xmax>0.59</xmax>\n <ymax>0.83</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00dc48131468b9f0.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Woman</name>\n <bndbox>\n <xmin>0.17</xmin>\n <ymin>0.12</ymin>\n <xmax>0.39</xmax>\n <ymax>0.46</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.55</ymin>\n <xmax>0.09</xmax>\n <ymax>0.64</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.14</ymin>\n <xmax>0.31</xmax>\n <ymax>0.18</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/072e68abf4f016ec.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bed</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.7</ymin>\n <xmax>0.96</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n <object>\n <name>Cat</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.35</ymin>\n <xmax>0.62</xmax>\n <ymax>0.8</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03981a436416648c.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cowboy hat</name>\n <bndbox>\n <xmin>0.13</xmin>\n <ymin>0.09</ymin>\n <xmax>0.75</xmax>\n <ymax>0.38</ymax>\n </bndbox>\n </object>\n <object>\n <name>Fedora</name>\n <bndbox>\n <xmin>0.09</xmin>\n <ymin>0.08</ymin>\n <xmax>0.76</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sun hat</name>\n <bndbox>\n <xmin>0.08</xmin>\n <ymin>0.08</ymin>\n <xmax>0.77</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0001e27f4b156f49.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.32</ymin>\n <xmax>0.47</xmax>\n <ymax>0.34</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.52</xmin>\n <ymin>0.31</ymin>\n <xmax>0.54</xmax>\n <ymax>0.33</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.31</ymin>\n <xmax>0.56</xmax>\n <ymax>0.32</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.69</xmin>\n <ymin>0.57</ymin>\n <xmax>0.69</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.7</xmin>\n <ymin>0.6</ymin>\n <xmax>0.72</xmax>\n <ymax>0.61</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human mouth</name>\n <bndbox>\n <xmin>0.47</xmin>\n <ymin>0.38</ymin>\n <xmax>0.48</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human mouth</name>\n <bndbox>\n <xmin>0.54</xmin>\n <ymin>0.36</ymin>\n <xmax>0.56</xmax>\n <ymax>0.37</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human mouth</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.63</ymin>\n <xmax>0.69</xmax>\n <ymax>0.66</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human body</name>\n <bndbox>\n <xmin>0.72</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Mammal</name>\n <bndbox>\n <xmin>0.33</xmin>\n <ymin>0.22</ymin>\n <xmax>0.51</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Mammal</name>\n <bndbox>\n <xmin>0.5</xmin>\n <ymin>0.07</ymin>\n <xmax>0.54</xmax>\n <ymax>0.15</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03a23823ea019ba6.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Table</name>\n <bndbox>\n <xmin>0.75</xmin>\n <ymin>0.76</ymin>\n <xmax>0.99</xmax>\n <ymax>0.97</ymax>\n </bndbox>\n </object>\n <object>\n <name>Chair</name>\n <bndbox>\n <xmin>0.44</xmin>\n <ymin>0.76</ymin>\n <xmax>0.57</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n <object>\n <name>Wheel</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.9</ymin>\n <xmax>0.16</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.17</xmin>\n <ymin>0.22</ymin>\n <xmax>0.33</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.34</xmin>\n <ymin>0.22</ymin>\n <xmax>0.46</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.64</xmin>\n <ymin>0.23</ymin>\n <xmax>0.79</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06f5e60505dec0c6.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Box</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0.1</ymin>\n <xmax>0.81</xmax>\n <ymax>0.86</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/077f5162b0fe17ec.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Furniture</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08954de74350e0a6.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.06</ymin>\n <xmax>0.49</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07c5c0c1b9f9cda1.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cosmetics</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.54</ymin>\n <xmax>0.55</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00078712936e5c4b.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Butterfly</name>\n <bndbox>\n <xmin>0.11</xmin>\n <ymin>0.08</ymin>\n <xmax>0.73</xmax>\n <ymax>0.33</ymax>\n </bndbox>\n </object>\n <object>\n <name>Butterfly</name>\n <bndbox>\n <xmin>0.16</xmin>\n <ymin>0.35</ymin>\n <xmax>0.71</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n <object>\n <name>Butterfly</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0.61</ymin>\n <xmax>0.68</xmax>\n <ymax>0.82</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02f322db7a5996b5.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cat</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.41</ymin>\n <xmax>0.56</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/015155a62fd65a13.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Pumpkin</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.49</ymin>\n <xmax>0.83</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n <object>\n <name>Pumpkin</name>\n <bndbox>\n <xmin>0.11</xmin>\n <ymin>0.05</ymin>\n <xmax>0.49</xmax>\n <ymax>0.54</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/036b8e5b2408f356.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.82</xmin>\n <ymin>0.79</ymin>\n <xmax>0.94</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0014c3b48667efde.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Toy</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.57</ymin>\n <xmax>0.39</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n <object>\n <name>Girl</name>\n <bndbox>\n <xmin>0.33</xmin>\n <ymin>0.1</ymin>\n <xmax>0.74</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/026ace2555b5f02d.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Spoon</name>\n <bndbox>\n <xmin>0.47</xmin>\n <ymin>0.17</ymin>\n <xmax>0.65</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n <object>\n <name>Spoon</name>\n <bndbox>\n <xmin>0.64</xmin>\n <ymin>0.17</ymin>\n <xmax>0.9</xmax>\n <ymax>0.6</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07c4c14d0b894f77.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.04</xmin>\n <ymin>0</ymin>\n <xmax>0.68</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bottle</name>\n <bndbox>\n <xmin>0.7</xmin>\n <ymin>0.79</ymin>\n <xmax>0.8</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0285f617f6cdf4ba.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.28</xmin>\n <ymin>0.29</ymin>\n <xmax>0.76</xmax>\n <ymax>0.82</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.09</ymin>\n <xmax>0.61</xmax>\n <ymax>0.24</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00421577770d9db5.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Drink</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.61</ymin>\n <xmax>0.97</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/022f06ac4b731813.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Crown</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.07</ymin>\n <xmax>0.82</xmax>\n <ymax>0.27</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/051233a32a0bbf2e.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Poster</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0316ff2baf27cf29.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Giraffe</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.15</ymin>\n <xmax>0.6</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0692df6384563bd0.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.16</xmin>\n <ymin>0.22</ymin>\n <xmax>0.31</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/045707a371224f49.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.57</ymin>\n <xmax>0.36</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>Building</name>\n <bndbox>\n <xmin>0.11</xmin>\n <ymin>0.09</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.69</ymin>\n <xmax>0.02</xmax>\n <ymax>0.76</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.55</ymin>\n <xmax>0.03</xmax>\n <ymax>0.6</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.85</ymin>\n <xmax>0.05</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.03</xmin>\n <ymin>0.68</ymin>\n <xmax>0.07</xmax>\n <ymax>0.75</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.53</ymin>\n <xmax>0.08</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.07</xmin>\n <ymin>0.84</ymin>\n <xmax>0.11</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.08</xmin>\n <ymin>0.65</ymin>\n <xmax>0.12</xmax>\n <ymax>0.73</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.1</xmin>\n <ymin>0.5</ymin>\n <xmax>0.14</xmax>\n <ymax>0.57</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.63</ymin>\n <xmax>0.22</xmax>\n <ymax>0.86</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.18</xmin>\n <ymin>0.37</ymin>\n <xmax>0.23</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0.6</ymin>\n <xmax>0.27</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0.33</ymin>\n <xmax>0.31</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.34</xmin>\n <ymin>0.28</ymin>\n <xmax>0.41</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.56</ymin>\n <xmax>0.4</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.29</ymin>\n <xmax>0.66</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.56</ymin>\n <xmax>0.65</xmax>\n <ymax>0.83</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.35</ymin>\n <xmax>0.72</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.68</xmin>\n <ymin>0.6</ymin>\n <xmax>0.73</xmax>\n <ymax>0.86</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.74</xmin>\n <ymin>0.4</ymin>\n <xmax>0.77</xmax>\n <ymax>0.57</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.74</xmin>\n <ymin>0.65</ymin>\n <xmax>0.79</xmax>\n <ymax>0.87</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.8</xmin>\n <ymin>0.51</ymin>\n <xmax>0.82</xmax>\n <ymax>0.63</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.81</xmin>\n <ymin>0.68</ymin>\n <xmax>0.84</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.85</xmin>\n <ymin>0.54</ymin>\n <xmax>0.87</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/030963627750d750.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.07</ymin>\n <xmax>0.75</xmax>\n <ymax>0.95</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0112a8b3ef79fa20.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Golf ball</name>\n <bndbox>\n <xmin>0.44</xmin>\n <ymin>0.59</ymin>\n <xmax>0.46</xmax>\n <ymax>0.62</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/023b74dc25cdbbe2.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Flower</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01047e9822575c85.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Pastry</name>\n <bndbox>\n <xmin>0.38</xmin>\n <ymin>0.41</ymin>\n <xmax>0.53</xmax>\n <ymax>0.65</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00c25ea2a49b6908.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.89</ymin>\n <xmax>0.4</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Building</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>0.93</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00d3d507e389f463.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bee</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.48</ymin>\n <xmax>0.7</xmax>\n <ymax>0.6</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/04e603dbe6e114fa.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.08</xmin>\n <ymin>0.25</ymin>\n <xmax>0.17</xmax>\n <ymax>0.64</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/044f24685a8f2a22.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Moths and butterflies</name>\n <bndbox>\n <xmin>0.17</xmin>\n <ymin>0.12</ymin>\n <xmax>0.83</xmax>\n <ymax>0.82</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07cb1062d1cc55f3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Jeans</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.71</ymin>\n <xmax>0.67</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tree</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.17</ymin>\n <xmax>0.7</xmax>\n <ymax>0.41</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.92</ymin>\n <xmax>0.62</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.62</xmin>\n <ymin>0.94</ymin>\n <xmax>0.64</xmax>\n <ymax>0.98</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02f9204ee8ad6507.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Boy</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.44</ymin>\n <xmax>0.61</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.58</ymin>\n <xmax>0.45</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.44</xmin>\n <ymin>0.59</ymin>\n <xmax>0.62</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.4</xmin>\n <ymin>0.47</ymin>\n <xmax>0.46</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/07b4cd46a385b4f8.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tart</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0296272c25074ec1.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Vehicle registration plate</name>\n <bndbox>\n <xmin>0.72</xmin>\n <ymin>0.37</ymin>\n <xmax>0.81</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0.12</ymin>\n <xmax>0.33</xmax>\n <ymax>0.3</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.12</ymin>\n <xmax>0.47</xmax>\n <ymax>0.31</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.89</xmin>\n <ymin>0.1</ymin>\n <xmax>1</xmax>\n <ymax>0.23</ymax>\n </bndbox>\n </object>\n <object>\n <name>Car</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0.05</ymin>\n <xmax>0.92</xmax>\n <ymax>0.92</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0539739ee5ce9be0.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Woman</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.54</ymin>\n <xmax>1</xmax>\n <ymax>0.97</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0418b55b2b74d429.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Suit</name>\n <bndbox>\n <xmin>0.01</xmin>\n <ymin>0.29</ymin>\n <xmax>0.2</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.03</xmin>\n <ymin>0.29</ymin>\n <xmax>0.11</xmax>\n <ymax>0.44</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.41</xmin>\n <ymin>0.34</ymin>\n <xmax>0.51</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.71</xmin>\n <ymin>0.33</ymin>\n <xmax>0.79</xmax>\n <ymax>0.5</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0296768e960a1e8a.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Footwear</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.09</ymin>\n <xmax>1</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03548fd47986828d.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.75</xmin>\n <ymin>0.76</ymin>\n <xmax>0.91</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>House</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.38</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.78</ymin>\n <xmax>0.38</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.67</ymin>\n <xmax>0.39</xmax>\n <ymax>0.77</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.39</xmin>\n <ymin>0.58</ymin>\n <xmax>0.45</xmax>\n <ymax>0.63</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.67</ymin>\n <xmax>0.53</xmax>\n <ymax>0.77</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.78</ymin>\n <xmax>0.54</xmax>\n <ymax>0.91</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.61</ymin>\n <xmax>0.68</xmax>\n <ymax>0.68</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.75</ymin>\n <xmax>0.68</xmax>\n <ymax>0.84</ymax>\n </bndbox>\n </object>\n <object>\n <name>Window</name>\n <bndbox>\n <xmin>0.71</xmin>\n <ymin>0.61</ymin>\n <xmax>0.9</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06269adab3d0dbfa.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.07</ymin>\n <xmax>0.44</xmax>\n <ymax>0.55</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.3</xmax>\n <ymax>0.74</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.3</ymin>\n <xmax>0.33</xmax>\n <ymax>0.57</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.33</xmin>\n <ymin>0.43</ymin>\n <xmax>0.37</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.44</xmin>\n <ymin>0.5</ymin>\n <xmax>0.47</xmax>\n <ymax>0.58</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.5</xmin>\n <ymin>0.5</ymin>\n <xmax>0.54</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.53</xmin>\n <ymin>0.44</ymin>\n <xmax>0.56</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.48</ymin>\n <xmax>0.62</xmax>\n <ymax>0.58</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.57</xmin>\n <ymin>0.32</ymin>\n <xmax>0.66</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.61</xmin>\n <ymin>0.41</ymin>\n <xmax>0.69</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.73</xmin>\n <ymin>0.37</ymin>\n <xmax>0.81</xmax>\n <ymax>0.5</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.77</xmin>\n <ymin>0.45</ymin>\n <xmax>0.81</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.81</xmin>\n <ymin>0.45</ymin>\n <xmax>0.86</xmax>\n <ymax>0.55</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.86</xmin>\n <ymin>0.44</ymin>\n <xmax>0.9</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tower</name>\n <bndbox>\n <xmin>0.9</xmin>\n <ymin>0.33</ymin>\n <xmax>1</xmax>\n <ymax>0.53</ymax>\n </bndbox>\n </object>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.07</ymin>\n <xmax>0.44</xmax>\n <ymax>0.59</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02553b1d3984ebff.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Pig</name>\n <bndbox>\n <xmin>0.17</xmin>\n <ymin>0.15</ymin>\n <xmax>0.74</xmax>\n <ymax>0.82</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/001ce300279f1a81.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Door</name>\n <bndbox>\n <xmin>0.9</xmin>\n <ymin>0.27</ymin>\n <xmax>1</xmax>\n <ymax>0.87</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/010db899406b65a6.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.01</xmin>\n <ymin>0.53</ymin>\n <xmax>0.03</xmax>\n <ymax>0.56</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.64</ymin>\n <xmax>0.05</xmax>\n <ymax>0.67</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.43</ymin>\n <xmax>0.08</xmax>\n <ymax>0.48</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.65</ymin>\n <xmax>0.08</xmax>\n <ymax>0.68</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.15</xmin>\n <ymin>0.45</ymin>\n <xmax>0.17</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.66</ymin>\n <xmax>0.44</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.45</xmin>\n <ymin>0.68</ymin>\n <xmax>0.47</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.52</xmin>\n <ymin>0.68</ymin>\n <xmax>0.55</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.56</xmin>\n <ymin>0.47</ymin>\n <xmax>0.6</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.59</xmin>\n <ymin>0.68</ymin>\n <xmax>0.61</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.87</xmin>\n <ymin>0.46</ymin>\n <xmax>0.9</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.89</xmin>\n <ymin>0.72</ymin>\n <xmax>0.91</xmax>\n <ymax>0.74</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sports equipment</name>\n <bndbox>\n <xmin>0.93</xmin>\n <ymin>0.68</ymin>\n <xmax>0.94</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0274165b127bd47e.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>House</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.31</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>House</name>\n <bndbox>\n <xmin>0.02</xmin>\n <ymin>0.04</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/044b37bca139f280.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Vehicle</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.52</ymin>\n <xmax>0.07</xmax>\n <ymax>0.6</ymax>\n </bndbox>\n </object>\n <object>\n <name>Vehicle</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0.27</ymin>\n <xmax>0.69</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02fe016b54c838f3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Fedora</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.18</ymin>\n <xmax>0.73</xmax>\n <ymax>0.38</ymax>\n </bndbox>\n </object>\n <object>\n <name>Sun hat</name>\n <bndbox>\n <xmin>0.49</xmin>\n <ymin>0.18</ymin>\n <xmax>0.73</xmax>\n <ymax>0.39</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/020294a97cbd1bba.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Bottle</name>\n <bndbox>\n <xmin>0.04</xmin>\n <ymin>0.04</ymin>\n <xmax>0.26</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bottle</name>\n <bndbox>\n <xmin>0.29</xmin>\n <ymin>0.13</ymin>\n <xmax>0.52</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bottle</name>\n <bndbox>\n <xmin>0.54</xmin>\n <ymin>0.27</ymin>\n <xmax>0.74</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n <object>\n <name>Bottle</name>\n <bndbox>\n <xmin>0.76</xmin>\n <ymin>0.4</ymin>\n <xmax>0.93</xmax>\n <ymax>0.96</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/048e97a148df0741.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.67</ymin>\n <xmax>0.1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.09</xmin>\n <ymin>0.7</ymin>\n <xmax>0.32</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.77</ymin>\n <xmax>0.51</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.57</xmin>\n <ymin>0.78</ymin>\n <xmax>0.79</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.97</xmin>\n <ymin>0.56</ymin>\n <xmax>1</xmax>\n <ymax>0.76</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/062d2386effbc304.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Wheel</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/02a940159cd39532.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.44</xmin>\n <ymin>0.32</ymin>\n <xmax>0.6</xmax>\n <ymax>0.37</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0</ymin>\n <xmax>0.9</xmax>\n <ymax>0.37</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human head</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0.01</ymin>\n <xmax>0.91</xmax>\n <ymax>0.65</ymax>\n </bndbox>\n </object>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human eye</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.3</ymin>\n <xmax>0.37</xmax>\n <ymax>0.35</ymax>\n </bndbox>\n </object>\n <object>\n <name>Shirt</name>\n <bndbox>\n <xmin>0.19</xmin>\n <ymin>0.47</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human mouth</name>\n <bndbox>\n <xmin>0.32</xmin>\n <ymin>0.46</ymin>\n <xmax>0.55</xmax>\n <ymax>0.55</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human ear</name>\n <bndbox>\n <xmin>0.78</xmin>\n <ymin>0.28</ymin>\n <xmax>0.92</xmax>\n <ymax>0.44</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.1</ymin>\n <xmax>0.79</xmax>\n <ymax>0.64</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human nose</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.33</ymin>\n <xmax>0.48</xmax>\n <ymax>0.45</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/03a6437d5e138df3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Jeans</name>\n <bndbox>\n <xmin>0.48</xmin>\n <ymin>0.01</ymin>\n <xmax>0.78</xmax>\n <ymax>0.69</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/01cfd918cf92072d.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.51</xmin>\n <ymin>0.47</ymin>\n <xmax>0.58</xmax>\n <ymax>0.61</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06f11bc57307517a.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.05</xmin>\n <ymin>0.73</ymin>\n <xmax>0.1</xmax>\n <ymax>0.85</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.68</ymin>\n <xmax>0.31</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.31</xmin>\n <ymin>0.7</ymin>\n <xmax>0.34</xmax>\n <ymax>0.8</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.37</xmin>\n <ymin>0.7</ymin>\n <xmax>0.41</xmax>\n <ymax>0.78</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.41</xmin>\n <ymin>0.77</ymin>\n <xmax>0.47</xmax>\n <ymax>0.89</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.77</ymin>\n <xmax>0.68</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.77</ymin>\n <xmax>0.68</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.72</xmin>\n <ymin>0.79</ymin>\n <xmax>0.8</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n <object>\n <name>Roller skates</name>\n <bndbox>\n <xmin>0.86</xmin>\n <ymin>0.8</ymin>\n <xmax>0.94</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0815654513ba23b3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Car</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.55</ymin>\n <xmax>0.2</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n <object>\n <name>Car</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.57</ymin>\n <xmax>0.42</xmax>\n <ymax>0.68</ymax>\n </bndbox>\n </object>\n <object>\n <name>Car</name>\n <bndbox>\n <xmin>0.77</xmin>\n <ymin>0.56</ymin>\n <xmax>1</xmax>\n <ymax>0.86</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0145aabe9dce2c5a.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.23</xmin>\n <ymin>0.48</ymin>\n <xmax>0.3</xmax>\n <ymax>0.73</ymax>\n </bndbox>\n </object>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.3</xmin>\n <ymin>0.46</ymin>\n <xmax>0.39</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.82</xmin>\n <ymin>0.57</ymin>\n <xmax>0.92</xmax>\n <ymax>0.73</ymax>\n </bndbox>\n </object>\n <object>\n <name>Skyscraper</name>\n <bndbox>\n <xmin>0.92</xmin>\n <ymin>0.61</ymin>\n <xmax>1</xmax>\n <ymax>0.73</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00d9a083e1cba929.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Screwdriver</name>\n <bndbox>\n <xmin>0.27</xmin>\n <ymin>0.45</ymin>\n <xmax>0.3</xmax>\n <ymax>0.9</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0042d5340087fb4d.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Girl</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0</ymin>\n <xmax>0.42</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/038cf4c2eb7ea83b.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Wine glass</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0</ymin>\n <xmax>0.7</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Wine glass</name>\n <bndbox>\n <xmin>0.8</xmin>\n <ymin>0.34</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/038596110c8e6992.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.64</ymin>\n <xmax>0.47</xmax>\n <ymax>0.79</ymax>\n </bndbox>\n </object>\n <object>\n <name>Table</name>\n <bndbox>\n <xmin>0.24</xmin>\n <ymin>0.56</ymin>\n <xmax>0.61</xmax>\n <ymax>0.8</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.09</xmin>\n <ymin>0.12</ymin>\n <xmax>0.14</xmax>\n <ymax>0.24</ymax>\n </bndbox>\n </object>\n <object>\n <name>Clothing</name>\n <bndbox>\n <xmin>0.18</xmin>\n <ymin>0.31</ymin>\n <xmax>0.25</xmax>\n <ymax>0.38</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.5</ymin>\n <xmax>0.04</xmax>\n <ymax>0.67</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.1</xmin>\n <ymin>0.1</ymin>\n <xmax>0.13</xmax>\n <ymax>0.14</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.2</xmin>\n <ymin>0.43</ymin>\n <xmax>0.24</xmax>\n <ymax>0.51</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.75</xmin>\n <ymin>0.26</ymin>\n <xmax>0.77</xmax>\n <ymax>0.29</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.88</xmin>\n <ymin>0.27</ymin>\n <xmax>0.9</xmax>\n <ymax>0.31</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/06a55a5c51371377.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Person</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.43</ymin>\n <xmax>0.7</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/00915674b39ba67c.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.08</xmin>\n <ymin>0</ymin>\n <xmax>0.64</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human hair</name>\n <bndbox>\n <xmin>0.87</xmin>\n <ymin>0.09</ymin>\n <xmax>1</xmax>\n <ymax>0.49</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/04e96e5104d98896.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Woman</name>\n <bndbox>\n <xmin>0.01</xmin>\n <ymin>0.41</ymin>\n <xmax>0.39</xmax>\n <ymax>0.99</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.06</xmin>\n <ymin>0.44</ymin>\n <xmax>0.19</xmax>\n <ymax>0.63</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/068bf4f046cc95e6.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.26</xmin>\n <ymin>0.07</ymin>\n <xmax>0.48</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.42</xmin>\n <ymin>0.15</ymin>\n <xmax>0.63</xmax>\n <ymax>0.52</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/026fd9ac9f553bd3.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Cattle</name>\n <bndbox>\n <xmin>0.14</xmin>\n <ymin>0.17</ymin>\n <xmax>0.66</xmax>\n <ymax>0.63</ymax>\n </bndbox>\n </object>\n <object>\n <name>Cattle</name>\n <bndbox>\n <xmin>0.82</xmin>\n <ymin>0.5</ymin>\n <xmax>1</xmax>\n <ymax>0.71</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/0458e49dbe451311.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Dress</name>\n <bndbox>\n <xmin>0.35</xmin>\n <ymin>0.42</ymin>\n <xmax>0.59</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Tiara</name>\n <bndbox>\n <xmin>0.4</xmin>\n <ymin>0.21</ymin>\n <xmax>0.47</xmax>\n <ymax>0.28</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/08cc19b1c57067bd.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Man</name>\n <bndbox>\n <xmin>0.25</xmin>\n <ymin>0</ymin>\n <xmax>1</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n <object>\n <name>Human face</name>\n <bndbox>\n <xmin>0.6</xmin>\n <ymin>0.04</ymin>\n <xmax>0.92</xmax>\n <ymax>0.57</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

[ { "content": [ { "path": null, "text": "You are an AI vision assistant specialized in object detection.\n\nThe user will provide exactly one image. \nYour task is to detect all visible, unoccluded objects in that image and output their categories and bounding boxes.\n\nOnly include objects that are **fully visible** — \ndo not annotate any object that is partially hidden, truncated, or occluded by other objects.\n\nFor each detected object, include:\n- <name>: the object category (e.g., \"person\", \"car\", \"dog\").\n- <bndbox>: the bounding box coordinates, using:\n <xmin>, <ymin>, <xmax>, <ymax>\n\nCoordinates must be **normalized** to the range [0, 1], \nrelative to the image width and height:\n- xmin = left / image_width \n- ymin = top / image_height \n- xmax = right / image_width \n- ymax = bottom / image_height\n\nYou must output **only** the results in the following **XML format**, with no explanations, no comments, and no extra text:\n\n```xml\n<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>person</name>\n <bndbox>\n <xmin>0.12</xmin>\n <ymin>0.08</ymin>\n <xmax>0.45</xmax>\n <ymax>0.72</ymax>\n </bndbox>\n </object>\n <object>\n <name>dog</name>\n <bndbox>\n <xmin>0.55</xmin>\n <ymin>0.35</ymin>\n <xmax>0.80</xmax>\n <ymax>0.88</ymax>\n </bndbox>\n </object>\n</annotation>\n```\n", "type": "text" } ], "role": "system" }, { "content": [ { "path": "/mnt/shared/p02/alex/fiftyone/open-images-v7/train/data/041b524627acae82.jpg", "text": null, "type": "image" } ], "role": "user" }, { "content": [ { "path": null, "text": "<?xml version=\"1.0\" ?>\n<annotation>\n <object>\n <name>Tiger</name>\n <bndbox>\n <xmin>0</xmin>\n <ymin>0.04</ymin>\n <xmax>0.99</xmax>\n <ymax>1</ymax>\n </bndbox>\n </object>\n</annotation>", "type": "text" } ], "role": "assistant" } ]

Datasets:

AlexHung29629
/

openimages_objdet

Models trained or fine-tuned on AlexHung29629/openimages_objdet

AlexHung29629/gpt-oss-20b-vl-sft-output-9