Workflow guide
One shoot, every feed, no crushed composition
Every platform wants a different rectangle, and naive center-cropping beheads speakers and strands captions under UI buttons. Reframing is a shot-by-shot craft of choosing what survives the crop. This guide covers the conversion paths from horizontal masters to vertical and square.
From the Mellius product team · Updated September 28, 2026
Conspiracy theorists, you may hate this
AI Director
›
Just type what you need
- ›Remove all silent parts
- ›Add bold kinetic captions
- ›Only have Bill Ackman speaking
- ›Cut to highlights only
- ›Add cinematic transitions
- ›Match cuts to the beat
The short answer
Duplicate the timeline per aspect, set the new frame size, reframe each shot around faces and action using position keyframes or tracking, keep text inside title-safe margins, then export 1080 by 1920 for vertical and 1080 by 1080 for square.
Subjects survive the crop
Faces, hands, and key action stay fully in frame in every aspect.
Graphics clear platform UI
Captions and titles avoid like, comment, and progress overlays on each platform.
Motion feels intentional
Reframing moves read as camerawork, not as drifting auto-crops.
Pick a conversion strategy per shot
Three options exist for each shot: center crop, manual reframe, or blurred-background fill. Center crop works only for centered subjects with headroom; it fails on two-shots, walk-and-talks, and graphics-heavy frames. Manual reframing with position keyframes is the default professional path — follow the speaker as they move, easing into new framings rather than snapping.
For wide shots that cannot survive a vertical crop, split the difference: punch into a medium shot for the vertical version, or build a 9:16 frame with the sharp subject over a blurred, darkened copy of the full frame. Never stretch or squeeze pixels to fit — viewers spot distorted faces instantly. In Mellius you can ask for a vertical reframe and then adjust any shot's framing on the timeline.
- Center crop only centered, static subjects.
- Keyframe position to follow moving speakers.
- Use blurred-fill backgrounds for uncroppable wides.
- Never anamorphically stretch footage to fit.
Protect graphics and export cleanly
Redesign the graphics layer per aspect instead of shrinking the horizontal layout. Move lower-thirds higher to clear TikTok and Reels side rails, enlarge caption type for phone viewing, and keep all text inside roughly 80% title-safe margins. Burned-in captions belong in the middle band of vertical frames where neither the top nor bottom UI reaches.
Export each aspect from its own timeline or sequence: 1920×1080 for YouTube, 1080×1920 for TikTok/Reels/Shorts, 1080×1080 for square feeds. Sharpen slightly for heavy-compression platforms and watch each export on a phone before shipping. Common mistake: exporting one 16:9 master and letting platforms auto-crop — their center crop will find your worst frame.
- Raise graphics above bottom UI on vertical video.
- Keep text inside 80% title-safe margins.
- Export separate masters per aspect ratio.
- Preview every version on a phone first.
A practical sequence
Put it into practice
- 01
Duplicate per aspect
Create separate timelines for 16:9, 9:16, and 1:1 instead of one compromised sequence.
- 02
Reframe shot by shot
Center, keyframe, or blurred-fill each shot so subjects and action survive.
- 03
Relayout and export
Reposition graphics clear of platform UI and export each size natively.
Frequently asked questions
How do I convert 16:9 video to 9:16 without losing quality?
Set a 1080×1920 sequence and reframe each shot with position and scale, which crops rather than stretches pixels. Shoot or master in 4K when possible so vertical crops retain detail. Expect to punch in: a vertical frame uses roughly the middle third of a horizontal shot.
Should I shoot horizontal or vertical for multi-platform?
Shoot horizontal 4K with vertical delivery in mind: center key action, leave headroom, and keep graphics flexible. That protects both versions. If one platform drives most revenue, shoot native for it and adapt outward rather than splitting the difference badly.
Why do my captions get covered on TikTok and Reels?
Platform UI occupies the right rail, bottom, and top of vertical video. Keep captions in the middle band inside title-safe margins and raise lower-thirds above the bottom overlay zone. Always preview the export inside the actual app before publishing.
Continue reading
Primary sources
Product capabilities change. These official references support the comparisons above.
Describe the edit. Keep the timeline.
Mellius turns conversational direction into cuts, captions, graphics, motion, color, and sound you can still inspect and change.
Start editing free