What pixelating a face does
Pixelation cuts the face into a grid of squares and paints each square one colour: the average of the pixels it replaces. With few enough squares, the eyes, nose and mouth merge into blocks that no longer show who the person is. The rest of the photo stays as it was.
This tool finds the faces for you with one face detection call, draws the mosaic in your browser and gives you a new JPG to download. The original file is not changed.
Choosing the block size
What hides a face is how many blocks fall across it, not how many pixels each block has. The tool sets the block size from each face, so every face it covers is hidden to the same degree, up to 100 faces in one photo. An area you draw by hand is treated as at least a third as tall as it is long, so a strip over a line of text gets blocks as large as the letters.
- Strong: about 6 blocks across a face. The face is a handful of flat squares.
- Medium: about 9 blocks across. Skin tone and hair colour still read; features do not.
- Light: about 14 blocks across. A stylised look. A face can stay recognisable at this setting, so don't rely on it for privacy.
Pixelate, blur or black box
Pixelation has hard edges, so a viewer can tell at a glance that a face was hidden on purpose. That is why newsrooms and screenshot tools use it. A blur is softer and blends into the photo; the face blur tool starts in that mode. The black box replaces the pixels with solid black and is the one to use when nothing of the face should remain. All three are a tap apart in the tool above.
Number plates, screens and missed faces
The detector looks for faces only. To pixelate anything else, such as a number plate, a house number, a phone screen or a signature, press Add an area and drag a box over it. Do the same for a face the tool missed because it was tiny, turned away or half hidden. Look over the photo before you share it.
Pixelate faces with the API
POST /v1/detect returns a box for every face in an image, as ratios of its width and height. Your code draws the mosaic over those boxes, on a server or in the browser. A call costs 93 credits per photo and stores nothing. The request and response are in the detect endpoint reference.
Frequently asked questions
- How do I pixelate a face in a photo?
- Add the photo. The tool finds the faces and covers each one with a mosaic of square blocks. Tap any face to switch it off or on, pick Light, Medium or Strong for the block size, and press Download. To cover a number plate, a screen or a face the tool missed, press Add an area and drag a box over it.
- How big should the blocks be?
- Big enough that the eyes, nose and mouth fall into just a few blocks. Strong puts about 6 blocks across a face, Medium about 9 and Light about 14. The block size follows each face, so a small face in the background gets small blocks and a close-up gets large ones.
- Is pixelating better than blurring?
- Both hide a face when they are strong enough. Pixelation has hard edges and reads as deliberate censoring, which suits news photos and screenshots. Blur is softer and draws less attention. The tool does both, and a solid black box for when nothing of the face should remain.
- Can pixelation be reversed?
- The original pixels can't be rebuilt from a few coarse blocks. Fine pixelation is weaker: with many small blocks a face can stay recognisable, and software has been trained to match finely pixelated faces. When it matters, use Strong or the black box.
- Are my photos stored?
- No. The photo is uploaded once so the faces can be found, processed in memory and discarded. The pixelation is drawn in your browser and the download is made on your device. No photo, face crop or faceprint is saved.
- What do I get when I download?
- A new JPG at the photo's own size, up to 4,096 pixels on the longer side. It is drawn in your browser, so the location and camera details stored in the original file are not copied into it.
- Is the face pixelator free?
- Yes. There is no sign-up and no card. A fair-use limit per visitor keeps the tool fast for everyone; if you reach it, the tool shows how long to wait.
- Can I pixelate faces automatically in my own product?
- Yes. POST /v1/detect in the SightRadar API returns a box for every face in an image. Your code draws the mosaic over those boxes, on a server or in the browser. Detect stores nothing.
More free tools
- Face compare Upload two photos and get a similarity score, a match verdict and a plain-English reading of it.
- Look-alike test Upload a photo of each person and see how much the two faces have in common.
- Mom or dad test Upload a photo of yourself and one of each parent to see whose face scores closer to yours.
- Face blur Upload a photo and the faces in it are blurred for you. Keep the ones you want sharp, then download.
- Face crop Upload a photo and get a centred crop of every face: square, portrait or circle, one file each.
- Face detector Upload a photo to count the faces in it, see a numbered box on each one, and check which are clear enough to match.
- API cost calculator Price your monthly face workload on AWS Rekognition, Azure Face, Google Cloud Vision and SightRadar, operation by operation.
- Rekognition calculator Price an AWS Rekognition image workload in any region, tier by tier: face search, detection, storage and liveness.