Remove the background from any image

Get a transparent PNG in seconds — people, products, and objects. AI that runs entirely in your browser.

Free · No signup · Files never leave your device

Drag and drop an image

click to browse, or paste an image (Ctrl+V)

How it works

1

Drop your image

Any PNG, JPG, or WebP photo — it stays on your device.

2

AI runs locally

A segmentation neural network runs right in your browser via WebGPU or WebAssembly. The model downloads once — nothing is ever uploaded.

3

Download your PNG

Get a transparent PNG for free, or pick a solid background color before downloading.

Background removal that never uploads your photos

Every other background remover works the same way: you upload your image to their server, their GPU processes it, and you trust them with the photo. This one flips that around — the neural network downloads to your browser once (~88 MB, cached), and from then on every image is processed on your own hardware. Nothing is uploaded, ever. That matters for photos of your kids, your ID badge shots, unreleased products, or anything under NDA.

The model, ISNet, was trained on people, products, animals, and everyday objects, and produces soft, accurate edges — including partial transparency where the subject blends into the background. The result is a standard transparent PNG that drops into Figma, Canva, PowerPoint, or any design tool; or pick a solid background color and the swap happens instantly, without re-running the AI.

Tips for clean cutouts

Good lighting and some contrast between subject and background give the sharpest mattes. Front-lit portraits, products on any reasonably distinct surface, and pets against grass or floors all cut out well. The hardest cases for any AI remover are wispy hair, fur, and glass — if an edge looks soft, try a photo with more subject-background contrast. You can paste images straight from the clipboard with Ctrl+V.

Frequently asked questions

Is it really free?

Yes. The background remover is completely free, with no signup, watermarks, or limits. We make money from our video translation service, not from this tool.

Do you upload my images?

No. The AI model runs locally in your browser using WebGPU or WebAssembly. Your images never leave your device — once the model has loaded, the tool even works offline.

What image formats are supported?

PNG, JPG, and WebP as input. The result is always a PNG — with a transparent background by default, or a solid color of your choice.

How accurate is it?

It uses ISNet, a dichotomous image segmentation network trained on people, products, animals, and everyday objects. Results are sharp for most photos; very fine details like wispy hair or fur are the hardest cases for any background remover.

Why does the first image take longer?

The AI model (~88 MB) is downloaded the first time you use the tool and cached by your browser. After that, processing starts instantly.

Can I use the results commercially?

Yes. They are your images, and the underlying AI model (ISNet) is Apache-2.0 licensed, so the output is free to use for any purpose, including commercial work.