README.md

metadata

library_name: transformers.js
pipeline_tag: object-detection
license: agpl-3.0

YOLOv10: Real-Time End-to-End Object Detection

ONNX weights for https://github.com/THU-MIG/yolov10.

Latency-accuracy trade-offs	Size-accuracy trade-offs

Usage (Transformers.js)

If you haven't already, you can install the Transformers.js JavaScript library from NPM using:

npm i @xenova/transformers

Example: Perform object-detection.

import { AutoModel, AutoProcessor, RawImage } from '@xenova/transformers';

// Load model
const model = await AutoModel.from_pretrained('onnx-community/yolov10l', {
    // quantized: false,    // (Optional) Use unquantized version.
})

// Load processor
const processor = await AutoProcessor.from_pretrained('onnx-community/yolov10l');

// Read image and run processor
const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/city-streets.jpg';
const image = await RawImage.read(url);
const { pixel_values } = await processor(image);

// Run object detection
const { output0 } = await model({ images: pixel_values });
const predictions = output0.tolist()[0];
const threshold = 0.5;
for (const [xmin, ymin, xmax, ymax, score, id] of predictions) {
    if (score < threshold) continue;
    const bbox = [xmin, ymin, xmax, ymax].map(x => x.toFixed(2)).join(', ')
    console.log(`Found "${model.config.id2label[id]}" at [${bbox}] with score ${score.toFixed(2)}.`)
}
// Found "person" at [473.05, 430.35, 533.53, 532.43] with score 0.92.
// Found "car" at [447.48, 378.60, 639.69, 478.38] with score 0.92.
// Found "person" at [549.94, 260.96, 591.81, 331.22] with score 0.91.
// Found "person" at [33.50, 469.62, 78.99, 571.88] with score 0.90.
// Found "car" at [177.90, 337.14, 399.34, 418.01] with score 0.90.
// Found "traffic light" at [208.80, 55.90, 233.13, 101.39] with score 0.90.
// Found "bicycle" at [449.02, 477.23, 555.98, 537.56] with score 0.89.
// Found "bicycle" at [352.45, 527.27, 463.67, 588.07] with score 0.89.
// ...