How to change a photo background to white
- Choose the photos (up to 50).
- Download the model the first time; Portrait suits people, Objects suits products.
- Select Remove background; the background turns white.
- Check the edges, adjust if needed and download the JPEGs.
Worked example
A selfie processed with the Portrait model keeps its width and height, with everything behind the person turned white; the result line shows what share of the image became background.
Formats and limits
| Input | JPEG, PNG, WebP |
|---|---|
| Output | JPEG, PNG |
| Limits | First use downloads a model: about 12 MB (Portrait) or 109 MB (Objects, needs WebGPU). Up to 40 megapixels and 50 images per batch. |
| Engine | transformers.js with MODNet (Apache-2.0, fp16) or BiRefNet-lite (MIT, fp16) in a Web Worker; WebGPU when available, otherwise WebAssembly (Portrait only) |
Limitations
- Hair, fur and see-through edges may need the Edge position and Feather controls.
- The Objects model needs WebGPU.
- Up to 40 megapixels per image and 50 images per batch.
- First use downloads the model (about 12 MB or 109 MB).
Questions
Which model should I use?
Portrait (about 12 MB) works for people. Objects (about 109 MB, needs WebGPU) handles products, animals and people.
Are my photos uploaded?
No. The model is downloaded once from Hugging Face and runs in your browser; the photos stay on your device.
How do I clean up the edges?
Use Edge position to trim halos or keep more of the subject, and Feather to soften the edge, then press Remove background again.
Privacy
On-device model. Processing runs in this browser. The open-source model files are downloaded once from the model host (Hugging Face) and cached; your content is not uploaded.
- Hugging Face: Only after you press Download model: requests for the model files (MODNet about 12 MB, or BiRefNet-lite about 109 MB) from huggingface.co and its CDN, which see your IP address. Your images are never sent.
This page opens the Background remover ready for this task. See the privacy policy for how toolsdocks handles data.