Posting a photo with other people in it
Event, classroom and street photos, where the people in the background did not agree to be posted. One pass covers every face it finds.
Blur every face in a photo automatically, or drag a box over a name, an address or a licence plate. Everything happens in your browser, and the saved copy carries no EXIF data — GPS included.
Drop a photo to blur what shouldn't be in it
Faces are found automatically. JPG, PNG, WebP or iPhone HEIC.
Most "blur an image online" sites ask you to upload the photo you are trying to keep private. That is a strange trade even when the site is honest: the picture you least want on someone else's server is the one with your address, your child's face, or your passport in it. This one loads a face-detection model into your browser and does the work there. The photo never leaves your device, and you can confirm that in the network panel while it runs.
Faces are found automatically and boxed the moment the photo lands, because in a group shot that is the part a person gets wrong — one face at the edge of the frame is easy to miss and impossible to un-post. Everything else you draw yourself: a name on a form, a street number, a licence plate, a message thread in a screenshot. The automatic boxes are editable like any other, so a false positive is one click to delete.
Three ways to cover: a blur, a mosaic, or a solid block. Each covers the full rectangle — nothing is left half-obscured to advertise an upgrade — and the saved copy is re-encoded from scratch, which strips every EXIF field including the GPS coordinates that would otherwise say exactly where the photo was taken.
Event, classroom and street photos, where the people in the background did not agree to be posted. One pass covers every face it finds.
Cover the account name, email address and order number before the screenshot goes into a ticket or a blog post.
Marketplace and rental listings: blur the licence plate, the house number and anything on the fridge door.
Black out an account number or a signature on a photographed form before you send it on.
Faces are detected on arrival and boxed automatically. Delete any box that isn't a face.
Drag on the photo to add a region over a name, an address or a plate. Choose blur, mosaic or a solid block.
Download at the original resolution as PNG or JPG, with all EXIF data removed.
No. The face-detection model is downloaded to your browser and the photo is processed there. Nothing is sent to a server at any point.
Not from the saved file. The pixels are replaced, not hidden behind a layer, so there is nothing underneath to recover — unlike drawing a box in a PDF or a document editor, where the original often survives underneath.
It finds most of them, and it is deliberately tuned to over-detect rather than miss. It is still a model: check the result, and add anything it missed by hand. That is why the boxes are editable.
The single-photo tool is completely free — full resolution, no watermark, no limit on the number of areas or faces. A privacy tool that half-covers a face to sell an upgrade would be worse than useless. Pro adds the batch pass.
It auto-blurs every face across a folder of photos and hands back a ZIP — for event, classroom and listing shoots where doing it by hand is the only alternative.
Yes. The export is re-encoded from the pixels, so no EXIF field survives — including GPS coordinates, camera serial number and timestamps.