How to blur a face in a video
Blurring a face in one photo is easy. In a video the face moves, so the cover has to follow it. Here are the options, and a way that never uploads the clip.
Why video is harder than a photo
A blur placed over a face in a photo stays where you put it. In a video, people move, turn and leave the frame, so a fixed blur misses the face within a second. The cover has to be found again in each frame. If you only have one still image, read how to blur a face in a photo instead.
On an iPhone
Apple's guide to cropping, rotating, flipping and straightening in the Photos app lists those tools and does not mention blurring a face, as of October 2026. Cropping is the only way to cut someone out there, and only if they are at the edge of the frame.
In Clipchamp on Windows
Microsoft's Clipchamp has a blur effect. Its own blog describes blurring one face with a workaround: crop a copy of the clip to the face, apply the blur effect to it, then lay the original underneath as picture-in-picture. The same article says this suits clips without much movement, and it describes no automatic face tracking. For a person who walks around, you would be adjusting by hand.
In the browser, without uploading
The blur faces in video tool finds faces itself and covers every one it finds in the whole video. There is no way to pick a single face, so everyone it sees is covered. It takes MP4, MOV, WebM and MKV videos, one or several at once.
- How it finds faces: about ten times a second it checks the frame with a face detector that runs on your device, follows each face between checks, and keeps it covered for half a second after it was last seen. The detector downloads once, about 4 MB, then is cached.
- Cover: blur, pixelate, a black box, or an emoji. Blur and pixelate have a strength from 1 to 10.
- Sound: kept unless you tick "Remove the sound too", worth doing when voices could identify someone.
- Result: an MP4. The video is decoded and encoded again, so the file size can change, and a note says in how many frames faces were covered.
In our test, a 2 second clip with a cartoon face drawn on it moving across the frame had the face covered in 100% of frames, and a clip with no face in it said "no faces found: check the video before sharing it". These were synthetic clips with no real people in them, so they show the tool runs and reports honestly, not how well it finds real faces in your footage.
Everything happens inside your browser tab, so the video never leaves your device. That matters most for exactly the clips you want to anonymize.
Check the result before you share it
A face detector can miss faces in profile, turned away, very small, or half hidden by a mask or glasses, even for a moment. Watch the whole result. A black box or an emoji hides a face completely, while a light blur may still let someone who knows the person recognize them. To also drop the soundtrack, use mute video.
Questions
Can I blur just one person's face in a video?
Not with the blur faces in video tool, which covers every face it detects. To leave one person visible you would need an editor where you place the blur by hand.
Does it follow a face that moves?
Yes. It looks for faces about ten times a second, follows them between checks and keeps each covered for half a second after it was last seen.
Will it catch every face?
No detector does. Faces in profile, turned away, very small or partly covered can be missed for a moment, so watch the result before you share it.
Is the video uploaded?
No. The video is read, scanned and written again by code in your browser tab, and the face detector runs on your device.