Images & Design
Remove plain backgrounds from multiple images at once
Upload JPG, PNG or WEBP product shots on a plain, near-solid backdrop. You get a transparent PNG and a mask for each. It does not work on busy or detailed backgrounds.
or drop them here
- JPG / PNG / WEBP
- Up to 20 images
Remove a border-connected plain background while retaining the centered foreground
- The corner color and tolerance are recorded
- Both foreground and transparent pixels are verified
- Complex scenes remain outside this conservative processor
source-01-product-cutout.png · source-01-product-mask.png · background-removal-evidence.csv · manifest.json
One clear job, from source to download
- 1
Add the source
Supported formats and limits are visible before the upload.
- 2
Confirm the settings
Review the exact source, options, units and access before processing.
- 3
Inspect and download
Check the preview and warnings, then unlock the complete package.
Cutting out product shots on a plain backdrop
What it works on
You upload up to 20 JPG, PNG or WebP images in one batch. The tool is made for product shots on a plain, near-solid backdrop. It removes only pixels that touch the image border and fall within a color tolerance. It does not detect the subject, so a product on a busy or detailed background will not be isolated. Use it where the backdrop is one even color.
What you get back
For each image you get a transparent PNG and a separate alpha mask. An evidence CSV lists the corner color and tolerance used per image, and a manifest records a fingerprint of each source. Every cutout is reopened to confirm that both foreground and transparent pixels remain. If either is missing, the job warns instead of returning an empty cutout. Look at the mask to judge the edges.
Tolerance and where it stops working
You choose the new background: transparent, white or a custom color. The border color tolerance can be off, light or strong. A smooth gradient backdrop often needs strong. At that setting, a product close in color to the backdrop can be eaten away. Hair, drop shadows and soft edges often keep part of the old background. Check every result before you use it in your shop.
Questions before you run it
Which image formats and how many files can I upload at once?
JPG, PNG and WEBP are accepted, up to 20 images in one batch. The limits are shown before the upload, so an oversized batch is rejected rather than partly processed.
Does it detect the subject automatically like an AI cutout tool?
No. It removes only pixels connected to the image border that fall inside the confirmed colour tolerance. There is no semantic segmentation, so a subject sitting on a busy photo background will not be isolated.
What happens to hair, shadows and soft edges?
Fine hair, drop shadows and soft edges usually keep part of the old background or lose foreground detail, because the cut follows colour distance rather than object shape. Every cutout ships with its alpha mask so you can inspect the edge before using it.
What is in the downloaded package besides the cutout?
Each image produces a transparent PNG and a separate alpha mask. Alongside them sits an evidence CSV listing the corner colour and tolerance used per image, plus a manifest with source hashes.
What if my product photo has a gradient or near-white background?
Raise the border colour tolerance and check the result: a smooth gradient often needs the strong setting, while a foreground close in colour to the background can be eaten away. If no opaque foreground or no transparent pixels remain after the pass, the job warns instead of returning an empty cutout.