onnx-community/yolov10m

Uso (Transformers.js)

// Si aún no lo ha hecho, puede instalar la biblioteca JavaScript Transformers.js de NPM usando:
npm i @xenova/transformers

// Ejemplo: Realizar detección de objetos.
import { AutoModel, AutoProcessor, RawImage } from '@xenova/transformers';

// Cargar modelo
const model = await AutoModel.from_pretrained('onnx-community/yolov10m', {
  // quantized: false,    // (Opcional) Usar versión no cuantizada.
});

// Cargar procesador
const processor = await AutoProcessor.from_pretrained('onnx-community/yolov10m');

// Leer imagen y ejecutar procesador
const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/city-streets.jpg';
const image = await RawImage.read(url);
const { pixel_values, reshaped_input_sizes } = await processor(image);

// Ejecutar detección de objetos
const { output0 } = await model({ images: pixel_values });
const predictions = output0.tolist()[0];

const threshold = 0.5;
const [newHeight, newWidth] = reshaped_input_sizes[0]; // Altura y anchura reformadas
const [xs, ys] = [image.width / newWidth, image.height / newHeight]; // Escalas de redimensionamiento x e y
for (const [xmin, ymin, xmax, ymax, score, id] of predictions) {
  if (score < threshold) continue;

  // Convertir a coordenadas originales de la imagen
  const bbox = [xmin * xs, ymin * ys, xmax * xs, ymax * ys].map(x => x.toFixed(2)).join(', ');
  console.log(`Found "${model.config.id2label[id]}" at [${bbox}] with score ${score.toFixed(2)}.`);
}
// Found "car" at [559.30, 472.72, 799.58, 598.15] with score 0.95.
// Found "car" at [221.91, 422.56, 498.09, 521.85] with score 0.94.
// Found "bicycle" at [1.59, 646.99, 137.72, 730.35] with score 0.92.
// Found "bicycle" at [561.25, 593.65, 695.01, 671.73] with score 0.91.
// Found "person" at [687.74, 324.93, 739.70, 415.04] with score 0.89.
// ...