Object Remover
Paint over an object and the area is refilled from the pixels around it. Works best for small objects on simple backgrounds.
No watermarkFree & unlimited
Best for simple backgrounds
Masked pixels are filled with an average of the pixels around them, so detailed backgrounds come back soft. For larger removals, work in several passes with a smaller brush.
All processing happens in your browser. No data is sent to any server.
About this tool
- 1
Upload your photo
Select a photo containing an unwanted object, person, or element.
- 2
Paint over the object
Use the brush to mask the area you want gone. Cover the whole object, plus a little of its edge.
- 3
Run the fill
Every masked pixel is replaced with a distance-weighted average of the unmasked pixels around it.
- 4
Refine and download
Paint over any leftover smudge and run it again, then save the cleaned photo as PNG.
- Paint a few pixels past the edge of the object so the fill has clean pixels to sample from.
- Small objects on uniform backgrounds - sky, grass, a painted wall - give the cleanest result.
- This is a pixel-averaging fill, not a generative model: it blurs rather than invents. Expect a soft patch on detailed backgrounds.
- Work one object at a time and re-run rather than masking everything in a single pass.
- Distance-weighted pixel fill
- Adjustable brush size
- Undo / redo on the mask
- Multi-pass refinement
- Runs entirely in your browser
- Remove a photobomber standing against a plain sky or wall.
- Clean up small distractions such as litter, a sign, or a power line.
- Erase dust spots and sensor blemishes from a scanned print.
- Tidy a product photo shot on a seamless backdrop.
Each masked pixel is replaced by a weighted average of the nearby unmasked pixels, with closer pixels counting for more. It is a classical inpainting-by-diffusion fill running on a canvas in your browser - there is no machine-learning model and nothing is uploaded.
Only if the background behind them is simple. Against a wall or sky the fill is convincing; against a crowd, foliage, or brickwork it will read as a blurred patch, because the tool can only reuse pixels that are already in the image.
On uniform backgrounds, yes. On detailed ones the filled area is noticeably softer than its surroundings. Zoom to 100% before you download, and keep masked areas small.
Related tools
View allMore in Images12
Updated