metadata
base_model: facebook/maskformer-resnet50-vistas
library_name: transformers.js
pipeline_tag: image-segmentation
https://huggingface.co/facebook/maskformer-resnet50-vistas 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 @huggingface/transformers
Example: Scene segmentation with onnx-community/maskformer-resnet50-vistas.
import { pipeline } from '@huggingface/transformers';
// Create an image segmentation pipeline
const segmenter = await pipeline('image-segmentation', 'onnx-community/maskformer-resnet50-vistas');
// Segment an image
const url = 'https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg';
const output = await segmenter(url);
console.log(output)
// [
// {
// score: 0.9999902844429016,
// label: 'Sky',
// mask: RawImage { ... }
// },
// {
// score: 0.9986440539360046,
// label: 'Terrain',
// mask: RawImage { ... }
// },
// ...
// }
// ]
You can visualize the outputs with:
for (let i = 0; i < output.length; ++i) {
const { mask, label } = output[i];
mask.save(`${label}-${i}.png`);
}
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).