Background Remover
A background remover isolates a subject by removing what is behind it. This one is not an AI model: you click the area you want gone — usually the background — and the tool flood-fills the connected region of similar colour from that point. A tolerance slider controls how similar a neighbouring pixel must be to be included, and you can click several regions in turn. Everything runs locally in your browser, so your image is never uploaded.
Overview
This tool takes a simple, predictable approach to background removal. Instead of an AI model guessing where the subject ends and the background begins, it uses region growing, also known as flood fill. You click a pixel, and the tool spreads outward from that point, absorbing every connected pixel whose colour is within the tolerance you have set. Click the background of a product photo shot against a plain surface and the whole backdrop goes in one action; click again on any patch that was missed. The advantage of this method is that it is honest and deterministic: you can see exactly which region will be removed, and the tolerance slider gives you direct control over how aggressive the selection is. It works well on flat, even backgrounds and much less well on busy ones, and no tool of this kind can cut cleanly around fine hair.
The technical core is a flood fill over the image data. The tool reads the pixels of your image into memory, measures the distance between the clicked colour and each neighbouring pixel, and expands the region while that difference stays inside the tolerance. It uses an explicit stack backed by a typed array rather than a queue of objects, and it yields to the event loop periodically, so the page keeps responding instead of freezing on large images even though the work happens on the main thread. There is no neural network, no training data and no segmentation model anywhere in the process, which also means the tool cannot infer a subject you have not clicked on. The compensating benefit is privacy and speed: no image, pixel or click ever leaves your device.
Because a flood fill follows connected colour, the quality of the result depends mostly on the source image. A uniform studio background, a plain wall or a flat colour block will come out cleanly; a gradient, a shadow or a textured surface will leave edges behind that you need to clean up with further clicks and a different tolerance. For subjects with soft or complicated edges, such as hair or fur, a click-based tool will always be less precise than a segmentation model, and it is worth knowing that before you start. What this tool does offer is a fast way to strip flat backgrounds from product photos, icons, screenshots and simple graphics, with no upload, no account and no waiting for a server round-trip.
Worked example
A 1500 × 1500 PNG product photo on a white sheet, cleared to transparency
A ceramics seller in Chiang Mai has a product photograph exported at 1500 × 1500 px as a PNG of 1.4 MB. The mug sits on a plain white sheet and was shot from directly above. The catalogue needs a transparent background so the mug can sit on a coloured page.
- Upload the PNG. It is drawn onto a 1500 × 1500 canvas, under the 16 megapixel cap, so no downscaling notice appears.
- Click a corner of the white sheet. The connected area of similar colour becomes transparent in a single pass. A click is not a brush: it grows the whole region it can reach from that pixel.
- At the default tolerance of 30 the sheet is gone but a light grey ring remains in the shadow under the handle.
- Raise the tolerance to 42. The fill is replayed from the untouched original automatically, so there is no need to click the image again.
- Check the body of the mug. The glaze is slightly off-white, so at 42 part of the highlight is inside the tolerance and turns transparent too. Drop back to 38, which clears the shadow and keeps the highlight.
- Click Download .png. The saved file is named mug-01-no-background.png.
Result: A 1500 × 1500 px PNG in which roughly 68 percent of the pixels are now transparent, saved back as PNG so the transparency survives. The size stays close to the original, because PNG is lossless, and the download keeps the .png extension taken from the source file.
How to use Background Remover
- Step 1: Choose an image. It is processed locally, so nothing is uploaded.
- Step 2: Click the background region you want removed. The connected area of similar colour is flood-filled from that point.
- Step 3: Adjust the tolerance slider and click again for any patches the first pass missed.
- Step 4: Download the result. Edges on textured or gradient backgrounds will need extra clicks.
Common mistakes
Clicking the background of a JPEG and downloading the result.
The download keeps the source format, and JPEG has no transparency, so the cleared area comes back black. Convert the photograph to PNG first, remove the background, and only then flatten it onto a colour in another editor if a JPEG is required.
Expecting a person, hair or fur to be cut out cleanly.
This is a flood fill: it removes a connected area of similar colour and stops wherever the colour changes. It suits flat, even backgrounds such as a paper sweep, a painted wall or a plain tablecloth. On hair or a busy background the edges will be ragged and a fringe will remain, and no slider setting fixes that.
Pushing the tolerance up to 150 to remove a shadowed background and finding holes in the subject.
Raise it in small steps and watch the subject. Anything connected to the clicked pixel and within the tolerance is cleared with it, so a white shirt against a white wall disappears too. Undo one click at a time, or lower the slider, which replays the fills from the original image.
Looking for a background colour or a replacement image.
This tool only makes pixels transparent. If the catalogue needs a specific colour behind the product, export the transparent PNG and place it on that colour in any editor that supports layers.
Features
- Click-to-remove flood fill
- Tolerance slider
- Multiple regions per image
- Runs locally, nothing uploaded
Limits worth knowing
- It is a click-to-select region-growing flood fill with a tolerance slider from 0 to 150, not a model that understands the picture. It handles flat, uniform backgrounds and fails on hair, fur, glass and busy backgrounds.
- Edges are hard. Anti-aliased pixels outside the tolerance stay, leaving a light fringe around the cut-out, and there is no feather, matting or brush tool for cleaning it up afterwards.
- The result is transparency in the source format. A JPEG source is re-encoded as JPEG at 0.92 quality and the cleared area becomes black, because JPEG cannot store alpha. Use PNG or WebP when the transparency has to survive.
- The canvas is capped at 16,000,000 pixels, so a larger photograph is downscaled before you can click it, and if reading the canvas fails the image is halved once more with a warning. There is no background replacement and no automatic subject detection.
- Undo removes one click at a time and every change replays all previous fills from the original, so a long click history gets slower as it grows. Everything runs on the device; no image or click coordinate leaves the browser. This is a Pro tool with a small free daily allowance.
When to use it
- E-commerce: Strip the plain backdrop from product photos so they sit cleanly on a white or branded background.
- Marketing: Remove the background from a flat graphic or logo for use in a banner or a social media post.
- Web design: Cut a screenshot or icon out of a solid-colour canvas without opening an image editor.
- Presentations: Isolate a simple graphic from a slide capture so it can be reused elsewhere.
- Personal use: Clean up a photo taken against a plain wall for a profile picture, keeping the original file on your own device.
Practical tips
- Start with a modest tolerance, around 20 to 40, and raise it gradually. Increasing it in small steps lets you stop at the point where the background is gone but the subject has not started to erode.
- Choose or shoot subjects against a plain, evenly lit background. A solid colour that contrasts with the subject is the single biggest factor in getting a clean result from a flood-fill tool.
- Click the background, not the subject. Each click removes the connected region of similar colour around the point you clicked, so clicking the subject by mistake will remove the subject.
- Expect to clean up edges on textured or gradient backgrounds. Several clicks at different tolerances will get you most of the way; fine detail such as hair needs a dedicated segmentation tool.
Why use this tool
- Remove a flat background in seconds: One click on a plain backdrop and the whole connected region is gone, which is far quicker than tracing the subject by hand.
- Stay in control with the tolerance slider: A low tolerance removes only near-identical pixels, while a high tolerance spreads further and clears a background whose colour varies slightly.
- Work on several regions in one session: Click the main background, then click again on any pockets that were missed, adjusting the tolerance between clicks as needed.
- Keep your images private: The pixels are read, edited and re-encoded in your own browser, so nothing is uploaded to a server and no account is required.
- Get honest results on the right kind of image: Flat, even backgrounds come out cleanly, and the tool makes no claim to handle complex edges that a flood fill cannot resolve.
When this is not the right tool
- The subject has hair, fur or a soft edge that has to survive the cut-out. — A dedicated cut-out service or a desktop editor with refine-edge and matting tools. A flood fill cannot separate a soft edge from a background of a similar colour.
- You need the product on a specific colour or inside a new scene. — Remove the background here, then place the transparent PNG on the background you want in any editor that supports layers.
- The background is only slightly uneven and a fringe keeps showing up. — The Passport Photo tool, which uses a similar fill but also blends leftover near-background pixels towards the target colour. It hides a fringe, and it is worth trying when the final image does not need to stay transparent.
Frequently asked questions
What is a background remover?
It is a tool that removes the background from an image so the subject can be placed on a new background or left transparent. This one does it by flood-filling the region you click on.
How does this tool work?
You click a point in the background and the tool removes the connected area of similar colour, spreading outward until the colour differs by more than the tolerance you set. There is no AI model involved.
What are the benefits of using it?
It is fast and predictable on flat backgrounds, you control the selection directly with the tolerance slider, and the whole process runs locally so your image is never uploaded.
Who is it for?
Anyone who needs to strip a plain background from a product shot, logo, screenshot or simple graphic. Images with busy gradients or fine hair need a segmentation-based tool instead.
Is it free and private?
Yes. It is free, requires no account, and the image is processed entirely in your browser: no pixel ever leaves your device.
Related tools in Image Tools
- Image Resizer — The Image Resizer lets you resize images to exact dimensions or by percentage.
- Image to PDF — Image to PDF converts one or multiple images into a single PDF document.
- Image Converter — Free Image Converter — convert between PNG, JPG, WebP, GIF, BMP, and TIFF formats instantly.
- Image Compressor — Free Image Compressor — reduce image file size by up to 95% while maintaining visual quality.
- Compress Image to Exact Size (KB) — Compress an image to an exact target file size in KB or MB.
- Screenshot Tool — The Screenshot Tool captures a still image of a screen, window, or tab that you choose in your browser picker.
- Passport Photo Maker — The Passport Photo Editor helps you crop and format photos to standard passport and visa dimensions for multiple countries.
- Batch Image Editor — A batch image editor is an indispensable tool for anyone working with large volumes of images, from professional photographers and e-commerce store ow
- Photo Enhancer — An AI Photo Enhancer improves how a photo looks.