Remove anything from a photo right in your browser — no upload, no sign up
Choose an image, draw a box around the object you want gone, then download the cleaned up PNG

Drag a box around the object you want to remove
100% Private, your images never leave your browser. The object is removed locally on your device. Nothing is uploaded to our servers.
This tool runs LaMa, a large‑mask inpainting model, running entirely inside your browser. Draw a box around the thing you want gone and the model paints in what would be behind it, matching the surrounding textures, lines and lighting, then hands you a clean PNG. Draw another box to remove more.
The first time you use it, the tool downloads the AI model (~56MB) to your browser and caches it, so later runs start instantly. Only the area around your box goes through the model, so even large photos stay sharp and each removal takes seconds on most machines.
Want to edit images automatically at scale? The Bannerbear API automates it for you. Plug it into your app or no‑code workflow and process thousands of images on autopilot.
An AI object remover deletes something you do not want from a photo, and fills the area with what could be behind it.
This is also called inpainting. The AI model looks at the area around the object, then paints new pixels that match the textures, lines and light of the photo.
People use it to remove:
The tool works best on small and medium objects in front of a simple background, such as sky, grass, sand, a wall or a floor.
You can remove an object in a few seconds. You do not need to install software or make an account.
The tool works with most photos. These are the limits.
| Computer | iPhone or iPad | |
|---|---|---|
| Input formats | JPG, PNG, WEBP | JPG, PNG, WEBP |
| Maximum file size | 60 MB | 25 MB |
| Maximum output size (longest side) | 4096 px | 2048 px |
| Output format | PNG | PNG |
The common questions about removing objects from photos, and their answers
No. The AI model runs in your browser, and your photo stays on your device. We do not see it, and we do not keep a copy.
Yes. The tool is free, with no watermark, no signup and no limit on the number of images.
Drag a box around each person, then click Remove Object. Include the shadow of the person in the box. For a group, remove one or two people at a time.
Yes. After each removal, the result stays in the editor. Draw a new box and click Remove Object again. To go back to the original photo, click Start over with the original image.
The box is too large, or the background behind the object has a lot of detail. Draw a smaller box, or remove the object in parts. The model is best with simple backgrounds, such as sky, grass or a wall.
The tool can remove small marks such as date stamps. Only remove a watermark from an image if you own it, or have permission to use it without the watermark.
The first time, your browser downloads the AI model (about 56 MB). The browser keeps it, so the next removal starts immediately.
Do you need to make hundreds of images for your products or content? The Bannerbear Image API makes each image from a template, with one API call.
This free tool is good for fixing one photo at a time. With Bannerbear, you design a template once. Then you send each photo URL and text from your app, or from a no-code tool such as Zapier or Make. Bannerbear can also remove the background of each photo, and make an AI background from a prompt.
Step-by-step guides:
Do you need to do more with your image? Use the background remover to make a transparent PNG, or the AI background changer to add a new backdrop.
Edit your images right in your browser, no upload and no watermark
Put your product on a colour or gradient backdrop with a drop shadow
Start a free trial with 30 API credits, no credit card required

