Paint over an unwanted object or distraction in your own photo and let AI reconstruct that area.
PNG, JPG, JPEG or WebP, up to 20 MB. The image is processed directly in your browser.
This tool uses inpainting: you indicate which area should disappear and the model tries to rebuild it from the content around it. It is designed to clean your own photographs, for example by removing distant people, cables, small objects or distracting elements.
Paint a mask directly over the photo.
Erase parts of the mask or start again without reloading the photo.
LaMa rebuilds the selected area using the context around.
The photo does not need to be sent to UtiliTools Pro servers.
Compare the original image with the generated result.
Download the edited image at its original resolution.
Select a PNG, JPG or WebP photo.
Paint over the entire object you want to remove.
Click “Remove object with AI” and wait for reconstruction to finish.
Compare the result and download the image in PNG or JPG.
The tool uses a LaMa model in ONNX format, run through WebAssembly. There is no per-image API fee. The first run downloads the model and may take longer than later runs.
Completely cover the object you want to remove and add a small margin around it. Small objects against relatively uniform backgrounds are usually reconstructed more successfully than large objects or complex geometric backgrounds.
Upload a photo, paint over the unwanted object or distraction, and AI reconstructs the area using the surrounding content.
It is not uploaded to UtiliTools Pro servers. The inpainting model is downloaded into your browser and processing is done locally on your device.
The first time the browser must download and prepare the artificial intelligence model. Then, the file can be stored in the browser cache.
You can use PNG, JPG, JPEG and WebP files up to 20 MB. To get better results you should fully mark the object and a small area around.
Yes. You can switch between the selection brush and mask eraser, adjust the brush size, or clear the entire selection.
No. Results depend on the object size, background complexity, and available context. Backgrounds with simple patterns are usually easier to reconstruct.
Separate the subject and create a transparent PNG.
Isolate an object with one click using SlimSAM.
Upscale a photo with AI super-resolution.
Adjust the crop of a photo.
Adjust sharpness, contrast and saturation.
Reduce your images’ file sizes after editing.
