image_id int64 458 3.53k | image imagewidth (px) 592 1.18k | width int32 592 1.18k | height int32 875 1.59k | objects listlengths 1 32 | detections listlengths 0 27 | raw_response stringlengths 2.91k 25k | inference_info stringlengths 644 646 |
|---|---|---|---|---|---|---|---|
3,337 | 964 | 1,221 | [
{
"bw_id": "5db0fa51ebaf18000100444b",
"category_id": 0,
"image_id": "3337",
"id": 1070,
"area": 113014,
"bbox": [
230,
52,
467,
242
],
"iscrowd": false
},
{
"bw_id": "12814",
"category_id": 6,
"image_id": "3337",
"id": 12814,
"area": 2... | [] | The user wants me to identify and classify distinct visual elements on a historical newspaper page.
**1. Analyze the input image:**
- It's a full page from the "Harrisburg Telegraph", dated October 20, 1917.
- The layout is dense with text columns, typical of early 20th-century newspapers.
- There are several... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... | |
2,346 | 834 | 1,036 | [
{
"bw_id": "5d8ff3e7ca9e6f00010031ff",
"category_id": 0,
"image_id": "2346",
"id": 3115,
"area": 25608,
"bbox": [
115,
36,
97,
264
],
"iscrowd": false
},
{
"bw_id": "48013",
"category_id": 6,
"image_id": "2346",
"id": 48013,
"area": 301... | [
{
"bbox": [
122,
36,
211,
246
],
"label": "Photograph",
"sub_label": "King Peter of Serbia"
},
{
"bbox": [
122,
262,
211,
363
],
"label": "Photograph",
"sub_label": "King Peter II"
},
{
"bbox": [
217,
36,
3... | The user wants me to identify and classify distinct visual elements on a historical newspaper page.
**1. Analyze the input image:**
- It's a full page from the "EVENING PUBLIC LEDGER—PHILADELPHIA, WEDNESDAY, JANUARY 2, 1918".
- The layout is dense with text columns, typical of early 20th-century newspapers.
-... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... | |
2,783 | 894 | 1,211 | [
{
"bw_id": "5dadc18ee8c9c80001001c0c",
"category_id": 0,
"image_id": "2783",
"id": 1344,
"area": 67155,
"bbox": [
269,
73,
121,
555
],
"iscrowd": false
},
{
"bw_id": "49558",
"category_id": 5,
"image_id": "2783",
"id": 49558,
"area": 30... | [
{
"bbox": [
13,
127,
94,
188
],
"label": "Illustration",
"sub_label": "Drawing of a bridge structure"
},
{
"bbox": [
188,
103,
483,
139
],
"label": "Headline",
"sub_label": "RAILROAD RUMBLES"
},
{
"bbox": [
188,
... | The user wants me to identify and classify visual elements from a scanned newspaper page.
1. **Analyze the Request:**
* **Input:** An image of a newspaper page (Harrisburg Telegraph, April 18, 1917).
* **Task:** Identify distinct visual elements.
* **Output Format:** JSON array of objects.
* *... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... | |
3,533 | 780 | 1,170 | [
{
"bw_id": "11889",
"category_id": 2,
"image_id": "3533",
"id": 11889,
"area": 91195,
"bbox": [
477,
59,
299,
305
],
"iscrowd": false
},
{
"bw_id": "11890",
"category_id": 6,
"image_id": "3533",
"id": 11890,
"area": 12870,
"bbox": [... | [
{
"bbox": [
114,
62,
236,
205
],
"label": "Advertisement",
"sub_label": "Subscription box for Roanoke Rapids Herald"
},
{
"bbox": [
309,
62,
439,
99
],
"label": "Photograph",
"sub_label": "Portraits of E. A. Matthews and T. W. Mason... | The user wants me to identify and classify distinct visual elements from the provided image of a historical newspaper page.
1. **Analyze the Image:** The image is a scanned page from a newspaper, likely from around 1918 based on the content (WWI references, flu epidemic mentions). It contains text columns, advertisem... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... | |
1,933 | 1,180 | 1,585 | [
{
"bw_id": "5d69a062f35043000100008d",
"category_id": 0,
"image_id": "1933",
"id": 3662,
"area": 74932,
"bbox": [
514,
954,
143,
524
],
"iscrowd": false
},
{
"bw_id": "5d4b4c3607638f0001001cc8",
"category_id": 0,
"image_id": "1933",
"id": 3... | [
{
"bbox": [
66,
295,
407,
417
],
"label": "Photograph",
"sub_label": "Soldiers in a dugout"
},
{
"bbox": [
66,
1119,
159,
1228
],
"label": "Illustration",
"sub_label": "Man standing next to shoe"
},
{
"bbox": [
66,
... | The user wants me to identify and classify visual elements from a provided image of a historical newspaper page.
1. **Analyze the Request:**
* **Input:** An image of a newspaper page ("The Ogden Standard", Dec 19, 1917).
* **Task:** Detect distinct visual elements.
* **Output Format:** A JSON array ... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... | |
2,345 | 991 | 1,219 | [
{
"bw_id": "5db5ab6aca9e6f00010046c6",
"category_id": 0,
"image_id": "2345",
"id": 872,
"area": 32596,
"bbox": [
384,
54,
116,
281
],
"iscrowd": false
},
{
"bw_id": "50634",
"category_id": 5,
"image_id": "2345",
"id": 50634,
"area": 110... | [
{
"bbox": [
16,
177,
129,
286
],
"label": "Photograph",
"sub_label": "Williams Grove Building"
},
{
"bbox": [
391,
177,
500,
286
],
"label": "Photograph",
"sub_label": "Portrait of a man in uniform"
},
{
"bbox": [
391,... | The user wants me to identify and classify visual elements on a historical newspaper page.
1. **Analyze the Request:**
* **Input:** An image of a newspaper page from January 30, 1918.
* **Task:** Detect distinct visual elements.
* **Output Format:** JSON array of objects.
* **Fields:** `bbox_2... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... | |
1,980 | 977 | 1,392 | [
{
"bw_id": "5b1823cca042dd00010014d0",
"category_id": 0,
"image_id": "1980",
"id": 5121,
"area": 63360,
"bbox": [
343,
176,
288,
220
],
"iscrowd": false
},
{
"bw_id": "38598",
"category_id": 5,
"image_id": "1980",
"id": 38598,
"area": 2... | [
{
"bbox": [
324,
22,
653,
46
],
"label": "Headline",
"sub_label": "Masthead: THE TOMAHAWK, WHITE EARTH, MINN."
},
{
"bbox": [
357,
106,
532,
160
],
"label": "Headline",
"sub_label": "Article title: IGNOTS WILD MEN OF THE MOUNTAIN"
... | The user wants me to identify visual elements in the provided image of a newspaper page.
1. **Analyze the image:** It's a scanned page from "The Tomahawk, White Earth, Minn." dated around 1918 (based on context clues like "World War", "Ignots", "German Stuff").
2. **Scan for distinct visual elements:**
* **Top... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... | |
878 | 703 | 1,031 | [
{
"bw_id": "5a0518b736def90001002b8e",
"category_id": 3,
"image_id": "878",
"id": 6157,
"area": 38372,
"bbox": [
411,
588,
212,
181
],
"iscrowd": false
},
{
"bw_id": "18616",
"category_id": 6,
"image_id": "878",
"id": 18616,
"area": 620... | [
{
"bbox": [
11,
47,
200,
88
],
"label": "Headline",
"sub_label": "Section title 'FARMERS'"
},
{
"bbox": [
415,
47,
636,
77
],
"label": "Headline",
"sub_label": "Article title 'Co-Operative Financing!'"
},
{
"bbox": [
4... | The user wants me to identify distinct visual elements on a historical newspaper page.
**1. Analyze the input image:**
- It's a scanned page from "The Sun, Price, Utah" dated June 1, 1917.
- The layout is typical of early 20th-century newspapers: dense columns of text, various headlines, some advertisements, and... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... | |
2,829 | 592 | 875 | [
{
"bw_id": "5b157068a042dd00010013cf",
"category_id": 0,
"image_id": "2829",
"id": 5020,
"area": 15900,
"bbox": [
33,
486,
100,
159
],
"iscrowd": false
},
{
"bw_id": "44897",
"category_id": 6,
"image_id": "2829",
"id": 44897,
"area": 23... | [
{
"bbox": [
31,
42,
130,
118
],
"label": "Illustration",
"sub_label": "Decorative graphic header reading 'FOR BETTER ROADS'"
},
{
"bbox": [
142,
42,
355,
92
],
"label": "Headline",
"sub_label": "Main article title 'Quality, Style, T... | The user wants me to identify and classify visual elements on a scanned newspaper page.
**1. Analyze the input image:**
- It's a single page from "THE KANSAS CITY NEWS".
- The layout is dense with text columns, typical of early 20th-century newspapers.
- There are several distinct sections:
- Top left: A... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... | |
458 | 929 | 1,234 | [
{
"bw_id": "5cf9d6a9aa800200010009e0",
"category_id": 3,
"image_id": "458",
"id": 4367,
"area": 88704,
"bbox": [
295,
71,
576,
154
],
"iscrowd": false
},
{
"bw_id": "11221",
"category_id": 6,
"image_id": "458",
"id": 11221,
"area": 7361... | [
{
"bbox": [
9,
49,
325,
74
],
"label": "Headline",
"sub_label": "KENTUCKY IRISH AMERICAN"
},
{
"bbox": [
9,
74,
325,
99
],
"label": "Headline",
"sub_label": "PUBLISHED EVERY SATURDAY..."
},
{
"bbox": [
604,
49,
... | The user wants me to identify and classify visual elements on a scanned newspaper page.
1. **Analyze the image:** It's a full page from the "Kentucky Irish American" newspaper, dated Saturday, December 13, 1917.
2. **Scan for distinct regions:** I need to look for headlines, articles, illustrations, cartoons, and ad... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... | |
3,134 | 855 | 1,119 | [
{
"bw_id": "5dd18c31ca9e6f00010059b9",
"category_id": 0,
"image_id": "3134",
"id": 1,
"area": 149298,
"bbox": [
45,
92,
447,
334
],
"iscrowd": false
}
] | [
{
"bbox": [
98,
29,
209,
84
],
"label": "Headline",
"sub_label": "Date box"
},
{
"bbox": [
274,
22,
611,
90
],
"label": "Headline",
"sub_label": "Newspaper Masthead"
},
{
"bbox": [
680,
28,
799,
84
... | The user wants me to identify and classify visual elements from a provided image of a historical newspaper page.
**1. Analyze the Request:**
* **Input:** An image of a newspaper page ("Evening Ledger", June 18, 1917).
* **Task:** Detect distinct visual elements.
* **Output Format:** JSON array of objects.
* **... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... | |
834 | 840 | 1,112 | [
{
"bw_id": "5d92681eebaf1800010033bd",
"category_id": 0,
"image_id": "834",
"id": 2797,
"area": 74816,
"bbox": [
238,
190,
112,
668
],
"iscrowd": false
},
{
"bw_id": "5d926662e8c9c80001000d4e",
"category_id": 1,
"image_id": "834",
"id": 280... | [
{
"bbox": [
207,
31,
633,
115
],
"label": "Headline",
"sub_label": "Masthead: Evening Illustrated Ledger"
},
{
"bbox": [
11,
118,
464,
185
],
"label": "Headline",
"sub_label": "Main Headline: POPE PROPOSES BASIS FOR UNIVERSAL PEACE"... | The user wants me to identify and classify visual elements from a provided image of a historical newspaper page.
1. **Analyze the Request:**
* **Input:** An image of a newspaper page ("Evening Ledger", August 14, 1917).
* **Output:** A JSON array containing objects for each distinct visual element.
* ... | {"model_id": "Qwen/Qwen3.6-35B-A3B", "prompt": "Detect every distinct visual element on this historical newspaper page. For each, return a JSON object with \"bbox_2d\": [x1, y1, x2, y2] (coordinates normalised to 0-1000), \"label\" (one of: Photograph, Illustration, Map, Comics/Cartoon, Editorial Cartoon, Headline, Adv... |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 37