No keyframing
Manually animating a crop across a whole clip is tedious and fragile. Auto reframe replaces it with a single automated pass.
Auto Reframe
Convert horizontal footage to vertical automatically: face tracking keeps the speaker centered while the frame crops to 9:16.
Workflow
How it works
Add horizontal footage
Import widescreen video — a podcast recording, an interview, a stream VOD shot at 16:9.
Track the subject
Face detection follows the speaker across the frame, so the crop knows where the action is at every moment.
Export at 9:16
The clip is re-composed vertically with the subject held in frame, ready for TikTok, Reels and Shorts.
Why it matters
What you get
Faces stay centered
The crop follows the detected face rather than locking to the middle of the frame, so speakers don't drift out of shot.
Built into clipping
Clips selected by the AI clip generator are reframed automatically — vertical output is the default, not an extra step.
Correctable when needed
If a shot needs a different framing, adjust it directly or describe what you want to the vibe editing agent.
FAQ
How does auto reframe decide what to keep in frame?
Face detection tracks the people in the shot and the crop follows them, keeping the active speaker inside the vertical frame.
What aspect ratios does it output?
The target is 9:16 vertical, the native ratio of TikTok, Instagram Reels and YouTube Shorts.
Does it work on videos without faces?
It is designed around people on camera. For footage without faces, you can set the framing manually before export.
Do I have to reframe clips one by one?
No. Reframing runs automatically on every clip the AI clip generator produces from a horizontal source.
Related
Start today
Create an account and use Auto Reframe when GPT-Video opens.
Plans from $19 a month. Cancel any day, keep the month.