Choose your image
Select a file, drag it onto the uploader, or paste an image from your clipboard. The browser validates the format, dimensions and memory requirements before processing.
Remove Image Backgrounds Instantly
Automatic background removal that runs directly in your browser. Your image stays on your device.
No image upload
Free to use
Detecting engine
Image background remover
Drop an image here
Choose, drag, or paste a photo. Background removal starts automatically after the image is validated.
PNG, JPG/JPEG or WebP · Up to 25 MB
Private by designYour image is processed locally in your browser and is not uploaded to BQTools. The processing model may be downloaded separately and cached by your browser.
Transparent
Removed
Original
Remove an image background online
BQ Background turns a PNG, JPG or WebP into a clean cut-out you can save as a transparent PNG or place on a new background.
Remove the background
The local processing model creates a soft alpha matte for the foreground. WebGPU is used when available, with WebAssembly compatibility mode as a fallback.
Download the result
Keep genuine transparency, choose a solid color, upload a backdrop, or blur the original background, then export a newly rendered PNG.
Private browser-based processing
Your source image is decoded, segmented and composited on your device. BQTools does not need an image upload endpoint to remove the background.
Your pixels stay local
The application does not send your source image, output, filename, EXIF data or segmentation mask to BQTools. Temporary image state is kept in browser memory and released when you replace or reset the project.
The model is a separate download
On first use your browser may download the pinned inference runtime and model files. Those requests are independent of your image and browser caching can avoid downloading large model assets again.
Transparent PNGs and useful background replacements
The checkerboard in the editor represents transparency only—it is never intentionally baked into a transparent PNG export.
Product photos
Cut a product away from a cluttered or white background, then export transparency for marketplaces, presentations, catalog layouts or another design tool.
Portraits and profile photos
Separate a person from the scene, compare the result with the original, and choose transparency, a clean solid color or your own backdrop.
Frequently asked questions
Answers about image privacy, transparency, formats, browser processing and the first model download.
Does BQ Background upload my images?
No. Source images are processed locally in your browser. The browser may download the processing runtime and model, but your image is not sent with those requests.
Is the background really transparent?
Yes. Transparent mode exports a PNG with alpha transparency. The gray checkerboard is only an editor preview and is not included in the transparent result.
Which image types can I use?
PNG, JPG/JPEG and WebP are accepted, up to 25 MB. Very high pixel-count images can also be rejected when processing them safely would require too much browser memory.
Why does the first background removal take longer?
The processing runtime and model need to be downloaded and initialized the first time. Browser caching can make later runs substantially faster, but cache persistence depends on browser storage policies.
Does it work without WebGPU?
Yes. The tool prefers WebGPU acceleration when available and initialization succeeds. If it cannot use WebGPU, it automatically retries using WebAssembly compatibility mode.
Can I change the background instead of making it transparent?
Yes. You can choose white, black, green, a custom color, upload another image, or blur the original scene. Those changes reuse the existing mask and do not rerun segmentation.
Does it work on phones?
The interface is designed for phones and tablets as well as desktop browsers. Available memory and processing speed vary by device, so very large images can require more conservative handling on mobile.
Are my images saved?
No user image is intentionally persisted by the app. Images are kept in temporary browser memory while you work. Browser storage is used for lightweight preferences and the processing runtime/model may use browser caching.