Update README.md

README.md

@@ -3,4 +3,191 @@ library_name: transformers.js
 tags:
 - pose-estimation
 license: agpl-3.0
----
+---
+
+YOLOv8x-pose with ONNX weights to be compatible with Transformers.js.
+
+## Usage (Transformers.js)
+
+If you haven't already, you can install the [Transformers.js](https://huggingface.co/docs/transformers.js) JavaScript library from [NPM](https://www.npmjs.com/package/@xenova/transformers) using:
+```bash
+npm i @xenova/transformers
+```
+
+**Example:** Perform pose-estimation w/ `Xenova/yolov8x-pose`.
+
+```js
+import { AutoModel, AutoProcessor, RawImage } from '@xenova/transformers';
+
+// Load model and processor
+const model_id = 'Xenova/yolov8x-pose';
+const model = await AutoModel.from_pretrained(model_id);
+const processor = await AutoProcessor.from_pretrained(model_id);
+
+// Read image and run processor
+const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/football-match.jpg';
+const image = await RawImage.read(url);
+const { pixel_values } = await processor(image);
+
+// Set thresholds
+const threshold = 0.3; // Remove detections with low confidence
+const iouThreshold = 0.5; // Used to remove duplicates
+const pointThreshold = 0.3; // Hide uncertain points
+
+// Predict bounding boxes and keypoints
+const { output0 } = await model({ images: pixel_values });
+
+// Post-process:
+const permuted = output0[0].transpose(1, 0);
+// `permuted` is a Tensor of shape [ 8400, 56 ]:
+// - 8400 potential detections
+// - 56 parameters for each box:
+//   - 4 for the bounding box dimensions (x-center, y-center, width, height)
+//   - 1 for the confidence score
+//   - 17 * 3 = 51 for the pose keypoints: 17 labels, each with (x, y, visibilitiy)
+
+// Example code to format it nicely:
+const results = [];
+const [scaledHeight, scaledWidth] = pixel_values.dims.slice(-2);
+for (const [xc, yc, w, h, score, ...keypoints] of permuted.tolist()) {
+    if (score < threshold) continue;
+
+    // Get pixel values, taking into account the original image size
+    const x1 = (xc - w / 2) / scaledWidth * image.width;
+    const y1 = (yc - h / 2) / scaledHeight * image.height;
+    const x2 = (xc + w / 2) / scaledWidth * image.width;
+    const y2 = (yc + h / 2) / scaledHeight * image.height;
+    results.push({ x1, x2, y1, y2, score, keypoints })
+}
+
+
+// Define helper functions
+function removeDuplicates(detections, iouThreshold) {
+    const filteredDetections = [];
+
+    for (const detection of detections) {
+        let isDuplicate = false;
+        let duplicateIndex = -1;
+        let maxIoU = 0;
+
+        for (let i = 0; i < filteredDetections.length; ++i) {
+            const filteredDetection = filteredDetections[i];
+            const iou = calculateIoU(detection, filteredDetection);
+            if (iou > iouThreshold) {
+                isDuplicate = true;
+                if (iou > maxIoU) {
+                    maxIoU = iou;
+                    duplicateIndex = i;
+                }
+            }
+        }
+
+        if (!isDuplicate) {
+            filteredDetections.push(detection);
+        } else if (duplicateIndex !== -1 && detection.score > filteredDetections[duplicateIndex].score) {
+            filteredDetections[duplicateIndex] = detection;
+        }
+    }
+
+    return filteredDetections;
+}
+
+function calculateIoU(detection1, detection2) {
+    const xOverlap = Math.max(0, Math.min(detection1.x2, detection2.x2) - Math.max(detection1.x1, detection2.x1));
+    const yOverlap = Math.max(0, Math.min(detection1.y2, detection2.y2) - Math.max(detection1.y1, detection2.y1));
+    const overlapArea = xOverlap * yOverlap;
+
+    const area1 = (detection1.x2 - detection1.x1) * (detection1.y2 - detection1.y1);
+    const area2 = (detection2.x2 - detection2.x1) * (detection2.y2 - detection2.y1);
+    const unionArea = area1 + area2 - overlapArea;
+
+    return overlapArea / unionArea;
+}
+
+const filteredResults = removeDuplicates(results, iouThreshold);
+
+// Display results
+for (const { x1, x2, y1, y2, score, keypoints } of filteredResults) {
+    console.log(`Found person at [${x1}, ${y1}, ${x2}, ${y2}] with score ${score.toFixed(3)}`)
+    for (let i = 0; i < keypoints.length; i += 3) {
+        const label = model.config.id2label[Math.floor(i / 3)];
+        const [x, y, point_score] = keypoints.slice(i, i + 3);
+        if (point_score < pointThreshold) continue;
+        console.log(`  - ${label}: (${x.toFixed(2)}, ${y.toFixed(2)}) with score ${point_score.toFixed(3)}`);
+    }
+}
+```
+
+<details>
+
+<summary>See example output</summary>
+
+```
+Found person at [535.7708740234375, 45.77457022666931, 644.4645690917969, 312.20427117347714] with score 0.697
+  - nose: (441.61, 87.47) with score 0.966
+  - left_eye: (449.36, 79.91) with score 0.988
+  - right_eye: (436.36, 79.56) with score 0.850
+  - left_ear: (462.02, 83.57) with score 0.919
+  - left_shoulder: (478.73, 127.16) with score 0.994
+  - right_shoulder: (420.37, 126.47) with score 0.703
+  - left_elbow: (503.33, 180.38) with score 0.977
+  - left_wrist: (506.53, 236.52) with score 0.924
+  - left_hip: (470.67, 223.60) with score 0.982
+  - right_hip: (432.32, 223.90) with score 0.851
+  - left_knee: (470.86, 306.20) with score 0.949
+  - right_knee: (428.56, 306.69) with score 0.601
+  - left_ankle: (463.92, 383.59) with score 0.737
+Found person at [-0.06377220153808594, 61.59769003391266, 156.24676704406738, 370.5519897222519] with score 0.926
+  - nose: (59.61, 100.49) with score 0.979
+  - left_eye: (66.44, 96.11) with score 0.954
+  - right_eye: (55.82, 96.21) with score 0.908
+  - left_ear: (76.90, 98.52) with score 0.819
+  - right_ear: (49.82, 102.11) with score 0.571
+  - left_shoulder: (87.07, 135.82) with score 0.990
+  - right_shoulder: (36.53, 134.96) with score 0.987
+  - left_elbow: (102.21, 193.66) with score 0.970
+  - right_elbow: (24.85, 187.30) with score 0.947
+  - left_wrist: (110.61, 245.75) with score 0.962
+  - right_wrist: (6.28, 233.46) with score 0.939
+  - left_hip: (82.71, 230.04) with score 0.997
+  - right_hip: (48.15, 235.65) with score 0.995
+  - left_knee: (95.27, 321.57) with score 0.993
+  - right_knee: (52.73, 320.56) with score 0.991
+  - left_ankle: (100.90, 415.89) with score 0.948
+  - right_ankle: (56.65, 417.09) with score 0.942
+Found person at [109.67742919921875, 12.466975402832032, 501.75636291503906, 533.3693368911744] with score 0.934
+  - nose: (126.43, 96.98) with score 0.715
+  - left_eye: (126.52, 88.36) with score 0.664
+  - left_ear: (136.92, 78.79) with score 0.934
+  - left_shoulder: (191.69, 125.31) with score 0.998
+  - right_shoulder: (166.08, 138.95) with score 0.993
+  - left_elbow: (254.38, 194.23) with score 0.997
+  - right_elbow: (186.09, 258.25) with score 0.986
+  - left_wrist: (309.75, 260.93) with score 0.990
+  - right_wrist: (133.20, 283.14) with score 0.973
+  - left_hip: (281.07, 280.72) with score 1.000
+  - right_hip: (258.20, 300.47) with score 1.000
+  - left_knee: (228.48, 442.67) with score 0.999
+  - right_knee: (250.90, 474.40) with score 0.999
+  - left_ankle: (343.96, 435.26) with score 0.979
+  - right_ankle: (340.41, 601.64) with score 0.971
+Found person at [422.38683700561523, 67.97338972091676, 638.0375099182129, 493.7016093254089] with score 0.932
+  - nose: (417.60, 144.74) with score 0.989
+  - left_eye: (426.67, 134.88) with score 0.959
+  - right_eye: (410.81, 135.93) with score 0.952
+  - left_ear: (443.39, 137.08) with score 0.771
+  - right_ear: (400.11, 142.05) with score 0.753
+  - left_shoulder: (446.92, 202.43) with score 0.997
+  - right_shoulder: (374.31, 196.36) with score 0.993
+  - left_elbow: (458.77, 287.40) with score 0.990
+  - right_elbow: (355.46, 260.60) with score 0.971
+  - left_wrist: (488.87, 354.68) with score 0.984
+  - right_wrist: (402.03, 263.57) with score 0.978
+  - left_hip: (432.69, 349.58) with score 0.998
+  - right_hip: (381.51, 366.30) with score 0.996
+  - left_knee: (463.97, 447.94) with score 0.991
+  - right_knee: (403.90, 511.95) with score 0.978
+  - left_ankle: (450.14, 562.29) with score 0.889
+  - right_ankle: (436.81, 548.29) with score 0.759
+```
+</details>