Restore and sharpen faces throughout a video. Faces are tracked from frame to frame so the result stays stable and flicker-free, and the original audio is kept.
The three largest faces in each frame are enhanced automatically — no masks or manual selection needed.
Face positions are smoothed across frames and scene cuts are detected, so restored faces stay steady instead of shimmering.
No. Only the faces are restored. Background, motion, timing and the original audio stay the same. Resolution follows the input, up to about 2000×1080, at up to 30 fps.
Up to 10 minutes.
$0.01 per second of video, billed in whole seconds with a 3-second minimum. The exact cost is shown before you generate, and new accounts get free credits to try it.