metadata
base_model: nvidia/segformer-b5-finetuned-cityscapes-1024-1024
library_name: transformers.js
pipeline_tag: image-segmentation
https://huggingface.co/nvidia/segformer-b5-finetuned-cityscapes-1024-1024 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: Image segmentation with Xenova/segformer-b5-finetuned-cityscapes-1024-1024.
import { pipeline } from '@xenova/transformers';
// Create an image segmentation pipeline
const segmenter = await pipeline('image-segmentation', 'Xenova/segformer-b5-finetuned-cityscapes-1024-1024');
// Segment an image
const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/cityscapes.png';
const output = await segmenter(url);
console.log(output)
// [
// {
// score: null,
// label: 'road',
// mask: RawImage { ... }
// },
// {
// score: null,
// label: 'sidewalk',
// mask: RawImage { ... }
// },
// ...
// ]
You can visualize the outputs with:
for (const l of output) {
l.mask.save(`${l.label}.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).