onnx-community/mediapipe_selfie_segmentation

Usage (Transformers.js)

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

npm i @huggingface/transformers

Example: Selfie segmentation with onnx-community/mediapipe_selfie_segmentation.

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

// Load model and processor
const model_id = 'onnx-community/mediapipe_selfie_segmentation';
const model = await AutoModel.from_pretrained(model_id, { dtype: 'fp32' });
const processor = await AutoProcessor.from_pretrained(model_id);

// Load image from URL
const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/selfie_segmentation.png';
const image = await RawImage.read(url);

// Pre-process image
const inputs = await processor(image);

// Predict alpha matte
const { alphas } = await model(inputs);

// Save output mask
const mask = await RawImage.fromTensor(alphas[0].mul(255).to('uint8')).resize(image.width, image.height);
mask.save('mask.png');

// (Optional) Apply mask to original image
const result = image.clone().putAlpha(mask);
result.save('result.png');

Input image	Predicted mask	Output image