How to crop a horizontal video to vertical and keep the face in frame
By the Cutora team · Last updated:
To turn a 16:9 talking-head video into 9:16, crop a vertical slice about a third of the frame wide and keep it centred on the speaker's face, then check every cut, because a speaker who leans or walks can drift toward the edge. In Cutora AI, a horizontal video exported vertical follows the face: each clip's crop is centred on the speaker's face when it finds one.
At a glance
- Horizontal to vertical
- Follows the face: one crop per clip, centred on the speaker's face when it finds one
- Vertical to horizontal
- Shown whole over a blurred, darkened copy of itself
- Export sizes
- 1080×1920 vertical or 1920×1080 horizontal
- Manual reframe control
- Not described in the product facts
Source: Cutora AI product facts. Last updated .
How do you do it?
How do I crop a horizontal video to vertical and keep the face in frame?
Crop a vertical slice about a third of the frame wide and keep it centred on the speaker's face, not on the middle of the frame. Then check each cut: the crop is set per clip, so a long clip in which you lean or walk can push you toward the edge. Recording with your head near the middle of the horizontal shot, and room above it, makes this easier.
Will cropping to vertical cut my head off?
It can, if the crop is centred on the frame and you sit off to one side. Centre the crop on your face instead, and look at each clip once before you post it. If the shot is wide on purpose, such as two people or a whiteboard, fit the whole frame over a blurred copy of itself rather than cropping.
How do I turn a 16:9 talking-head video into 9:16 for Shorts, Reels or TikTok?
Export it at 1080×1920 with the crop centred on the face, and keep captions away from the top and bottom edges, where the platforms put their buttons. Watch the clip once from start to end, because the crop is chosen per clip and a moving speaker can reach the edge.
What does Cutora do here?
Does Cutora keep my face in frame when it makes a vertical version?
Cutora follows the face. When a horizontal video is exported vertical (1080×1920), it centres each clip's crop on the speaker's face when it finds one, so the crop follows the face from clip to clip (one crop per clip, so a speaker who moves a lot inside a long clip can still reach the edge of the frame). A vertical video exported horizontal (1920×1080) is shown whole over a blurred, darkened copy of itself instead of being cut to a narrow strip. Check each clip before you post it.
Can I reframe by hand in Cutora?
Cutora's product facts do not describe a manual reframe control. After the first cut you can remove or restore any single cut from the Edits list, undo it, or ask the chat for a change; whether the chat can reframe is not stated.
Does the same crop apply to the Shorts Cutora cuts?
The Publish tab cuts Shorts from the video after the first cut (Lite 3, Creator 5, Pro 10, Studio 20 a month), and any horizontal video exported vertical gets the face-following crop. Cutora does not publish to social platforms: you post the clips yourself, and you should still watch each one first.
What does it not do?
Can Cutora track several speakers in a vertical crop?
No. Cutora AI is built for one person talking to camera, and it does not tell speakers apart. Its pages state no accuracy figure for finding the face, so look at each clip before you post it.