Blur Faces in Image
Faces are detected automatically and blurred or pixelated, so you can publish a photo of a crowd, a classroom or a street without exposing the people in it. Detection runs on the server with OpenCV, and the temporary processing workspace is removed after processing. It is free, needs no account, and adds no watermark to your result.
Blur Faces in Image
Add your files. Recommended settings are ready, so you can process in one click.
PDF, Word, images, audio, video, archives and more
Up to 100 files · 250 MB · No account- 1Add files
- 2Choose an action
- 3Download or keep going
What people use it for.
What happens to your file.
Haar cascade classifiers look for faces seen from the front and in profile, facing either way. Each region is grown past the detected box, which stops at the eyebrows and chin, and blurred or pixelated with a strength that scales with the size of the face, so a large face is hidden as thoroughly as a small one. The rest of the image is untouched, so the photo still reads normally. A solid black box is available too, the redaction newsrooms use. A PNG or WebP, such as a screenshot or a cut-out on a transparent background, stays in its format with its transparency, and the result never carries the original’s location or camera details.
Where it stops: Detection can still miss faces that are tilted sharply, turned away, heavily shadowed, covered or very small in frame. Always check the result before publishing, and use the manual blur tool for anything missed.
Everything you need, without a paywall.
Common questions.
What if a face is not detected?
Sharply tilted, turned-away and very small faces are the ones most often missed. Review the output, and use the Blur / Pixelate Image tool to cover anything the detector did not find.
Can the blur be reversed?
No. The pixels are destroyed rather than covered, so the original detail is not recoverable from the output file.
Can I use a black box instead of a blur?
Yes. Choose box as the mode for a solid black box over each face, the way newsrooms redact; nothing of the face remains in the pixels.
Should I remove the photo metadata too?
The result carries none of the original’s metadata, so its GPS location and camera details are already gone. If you publish the original anywhere, remove its metadata with Remove Image Metadata first.
You might also need.
Blur or pixelate selected parts of an image to hide faces, names, addresses, license plates and private information.
Strip EXIF, GPS and common image metadata while retaining image dimensions and visible content. Choose PNG, JPEG or WebP output.
Inspect image dimensions, format, EXIF fields, camera data, orientation, DPI and embedded metadata as JSON.
Permanently redact selected text or rectangular regions from PDF pages and burn the redaction into the saved output.