Xenova/yolov9-c

https://github.com/WongKinYiu/yolov9 with ONNX weights to be compatible with Transformers.js.

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 with Xenova/yolov9-c.

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

// Load model
const model = await AutoModel.from_pretrained('Xenova/yolov9-c', {
    // quantized: false,    // (Optional) Use unquantized version.
})

// Load processor
const processor = await AutoProcessor.from_pretrained('Xenova/yolov9-c');
// processor.feature_extractor.do_resize = false;                   // (Optional) Disable resizing
// processor.feature_extractor.size = { width: 128, height: 128 }   // (Optional) Update resize value

// 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 { outputs } = await model({ images: pixel_values })
const predictions = outputs.tolist();

for (const [xmin, ymin, xmax, ymax, score, id] of predictions) {
    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 "car" at [176.86, 335.53, 399.82, 418.13] with score 0.94.
// Found "car" at [447.50, 378.46, 639.81, 477.57] with score 0.93.
// Found "bicycle" at [351.90, 527.82, 463.50, 587.76] with score 0.90.
// Found "person" at [472.44, 430.52, 533.74, 533.30] with score 0.89.
// Found "bicycle" at [448.97, 477.34, 555.42, 537.63] with score 0.88.
// Found "bicycle" at [0.59, 518.69, 109.53, 584.31] with score 0.88.
// Found "traffic light" at [208.55, 55.80, 233.99, 101.63] with score 0.86.
// Found "person" at [550.75, 260.98, 591.90, 331.24] with score 0.86.
// ...

Demo

Test it out here!

---

Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using 🤗 Optimum and structuring your repo like this one (with ONNX weights located in a subfolder named onnx).