Free Image Eraser
Image Eraser removes objects, people, text or blemishes: paint over them or outline them with the lasso, and the MI-GAN AI model fills the gap. Only the area around what you painted is processed, and the picture keeps its original resolution. Erase in several passes, undo up to 12 of them and compare with the original.
- Tools:
- Brush, lasso and unmask
- Model:
- MI-GAN, 28 MB download
- Max size:
- Up to 40 MP, depending on memory
- Price:
- Free
- Runs:
- AI model in your browser
- Uploads:
- None: files never leave your device
- Account:
- Not needed
Quick answers
Is my picture uploaded?
No. The AI model runs inside your browser and your picture never leaves your device; only the model file is downloaded once from Hugging Face.
Does the rest of the photo lose quality?
No. Only the area around what you painted is processed. Every other pixel is kept exactly as it was, at the original resolution.
Processed on your device: your files are never uploaded. How it works
Network activity in this tab
What this page has requested since you opened the site in this tab. An upload is any request that sends data: POST, PUT, PATCH or a request with a body.
No uploads: nothing has been sent from this page.
No network requests yet.
To see every request, open your browser's developer tools (Network tab).
How to use the Image Eraser
- Drop the picture you want to clean up.
- Paint over the object, person, text or blemish with the brush, or draw around it with the lasso.
- Press Erase: the first time the AI model (28 MB) is downloaded, then the gap is filled in a moment.
- Repeat on other spots, compare with the original and download the full-resolution result.
Credits & open source
Open-source AI model running in your browser. Thank you to the people who make these projects.
- ONNX Runtime WebExecutes ONNX models with WebGPU or WebAssembly (self-hosted from /vendor/ort). © Microsoft Corporation · MIT
Web platform APIs used: Web Workers, WebAssembly, WebGPU, Canvas 2D, createImageBitmap, Pointer Events, Keyboard Events, File API.
- MI-GAN inpainting pipeline (ONNX) by Andranik Sargsyan, Shant Navasardyan, Xingqian Xu, Humphrey Shi (Picsart AI Research) (MIT) — AI model that fills the erased area
Full license texts: third-party notices · all credits
Frequently asked questions
How do I get the best result?
Cover the whole object, including its shadow, with a small margin. For big objects, erase in a few smaller passes; you can undo any pass.
Can it remove text and watermarks?
It can remove text, captions, date stamps and logos from your own pictures. Please only remove watermarks from images you have the right to edit.