Product photo background remover

Amazon requires a pure white background on the main image and rejects listings that do not comply. Etsy and Google Shopping do not require it, but clean, consistent backgrounds are the single biggest visual difference between a store that looks like a brand and one that looks like a garage sale. A transparent cut-out also lets you drop the product onto lifestyle scenes, ads and collection banners without reshooting.

This tool uses an open-source segmentation model that was trained to separate a subject from its surroundings. It handles most product shots, people, pets and objects with clear edges. Fine hair, glass, and thin straps are the hard cases; check those edges at full size before you publish.

Most background removers send your photo to a server, process it there, and keep a copy. This one downloads the model to your browser instead, then runs it on your own device using WebGPU when your browser supports it and a WebAssembly fallback when it does not. The model file is cached after the first visit, so later sessions start in a second or two.

The trade-off is speed on older hardware. A modern laptop processes an image in one to three seconds; a five-year-old phone may take ten or more. Nothing is sent anywhere, there is no limit, and there is no watermark.

Shoot on a plain background with the product clearly separated from it, and light the product evenly so its edges are not lost in shadow. If the model keeps part of the background, try again with a tighter crop using the image resizer first. For white backgrounds, choose white here rather than relying on the original photo, so every listing matches exactly.