Blur Faces in Photos, Free nothing stored

Find faces across the group, then check the count

Automatic detection looks at the whole photo and overlapping parts of it to look for smaller faces. The tool covers every face it detects with the same selected effect and reports how many it hid. You do not have to draw an outline around each person.

Compare that number with the people in the picture, then inspect their faces individually. A matching count is useful, but it does not prove the coverage is correct: a face-like object can be detected while a real person is missed.

Choose how much facial detail remains

The Effect choices are Blur, Pixelate or Black oval, with Blur selected initially. Softening blends nearby colors; a mosaic replaces patches with their average color; a solid cover replaces the color inside the detected oval with black.

The Strength control starts at 60. Raising it increases blur radius or mosaic block size relative to each face. It does not make the solid cover darker or enlarge the detected oval. Review the face boundary as well as the center.

Look for distant faces and uncovered profiles

Small background faces, turned heads and partly covered faces need a close look. The extra scans help search for them but cannot guarantee that everyone is found. If nothing is detected, the tool asks for a photo with clearer faces.

There is no manual box or brush here to cover a missed person. Try a closer crop in another editor, or cover the missed area there before sharing. Increasing effect strength changes an existing mask; it does not repair a missing detection.

Review more than faces in a school or street photo

For children’s photos, protest photos or bystanders in a street scene, inspect the finished image before posting. Check the edges of each covering, reflections and people at the back of the scene. Share the edited file, keeping the original separate.

Face hiding does not cover names on clothing, school signs, badges or license plates. It also cannot promise that someone is unrecognizable from their surroundings. If those details matter, crop them away or cover them in an editor with manual redaction.

Keep the unhidden photo on your device

The browser loads the face model and processing code, then reads and edits your picture locally. Your photo is not sent to a server for face detection or hiding. The face detector needs a model download, and other page requests also use the network.

Face hiding needs a phone, tablet or desktop browser that can run the detection model. Large images may be reduced to fit the device’s canvas limits. The result is a newly encoded picture, so its file size or file type can change. Check the saved copy before sending it.

Keep people covered or change the scene itself

This tool leaves people in the scene and covers detected faces. It does not remove their bodies, replace their identities or let you describe a different background. It also cannot leave a chosen person visible while hiding everyone else.

For a creative change, the linked Edit Image tool accepts instructions to modify picture content. Use it to describe a replacement or another edit. Generated changes are a different job from covering faces and are not a guarantee of anonymity.

How do you use the face blur?

1. Open the group picture

   Choose a JPG, PNG, WebP or HEIC photo and note the people whose faces need covering.

2. Set the face covering

   Choose from Blur, Pixelate or Black oval, then adjust Strength for the amount of softening or pixelation you need.

3. Run and inspect each person

   Start face hiding, compare before and after, and check the reported face count against the picture.

4. Save your result

   2 free downloads, 8 more after sign-in.

When is the free face blur enough?

Face Blur: questions and answers