Protect the subject
Avoid losing the speaker when widescreen footage becomes vertical.
Video intelligence
Change the frame.
Keep the subject visible.
Anjin uses detected face positions to plan portrait crops and validates whether a crop safely contains the people who need to remain on screen.
One video or an entire archive.
Face-aware reframing product demonstration
Visual direction preview
Visual direction preview9:16 · PREFLIGHT PASSED
FACE 0.94SOURCE 16:9FOLLOW FACEPRESERVE WIDEFALLBACK 1
ACTIVE SPEAKER
OUTPUT 9:1612visual decisions
94%crop confidence
1safe fallback
What is face-aware video reframing?
Aspect-ratio conversion guided by detected face position rather than a fixed centre crop.
The render pipeline stores face-crop data, validates crop coverage and refuses unsafe multi-face overlaps. This is face detection for framing, not facial identity recognition.
Controls and evidence
See what the system used and produced.
Useful automation leaves something a person can inspect. Anjin keeps the edit connected to its brief, source media and output settings.
FACE-AWARE REFRAMINGVERIFIABLE OUTPUT
- face_crop
- Detected face positions sampled across the source video.
- crop_mode
- The selected bars or supported stacked portrait treatment.
- coverage
- The proportion of the timeline with an admitted crop window.
- fallback
- Measured duration where safe face-aware cropping was unavailable.
Why it matters
More control, less timeline work.
Handle uncertainty
Use validation and fallback behaviour instead of pretending every crop is safe.
Inspect the result
Carry crop coverage information into render quality checks.
Fit and boundaries
Use the capability where it genuinely helps.
Anjin is strongest on speech-led recorded material where the editorial job can be described clearly and every selected moment needs to remain verifiable. It supports human judgement with searchable evidence, a reviewable plan and structured outputs.
A strong fit
Recorded knowledge with a story inside it
Interviews, podcasts, webinars, discussions and archive programmes benefit when the useful material is distributed across a long recording or several selected sessions.
Keep elsewhere
Final craft and image-led montage
Detailed colour, sound design, motion graphics and wordless visual storytelling remain finishing tasks for a professional editor and their preferred creative tools.
Evaluation checklist
What should you verify before adopting face-aware reframing?
face_crop
Confirm that the workflow exposes this clearly: detected face positions sampled across the source video.
crop_mode
Confirm that the workflow exposes this clearly: the selected bars or supported stacked portrait treatment.
coverage
Confirm that the workflow exposes this clearly: the proportion of the timeline with an admitted crop window.
fallback
Confirm that the workflow exposes this clearly: measured duration where safe face-aware cropping was unavailable.
Put it into practice
Start with a complete editing brief.
Questions about face-aware reframing
- Does face-aware reframing recognise who a person is?
- No. It detects face position for framing. It does not claim to identify the real-world identity of a person.
- What happens when several faces are visible?
- The crop validator checks whether the required faces can be safely included. Unsafe overlapping requirements can be refused rather than silently producing a poor frame.
- Which output is this most useful for?
- It is especially useful when converting widescreen interviews, podcasts and events into 9:16 portrait video.
- Can Anjin use a fixed frame instead?
- Yes. Supported crop settings and fallback output allow a workflow to prioritise predictable framing where needed.
Use face-aware reframing in your next edit.
Start with footage you already have and a clear description of the result you need.
One video or an entire archive.