Training, formats and accessibility · Ready to copy
Reframe a landscape interview for 9:16.
Plan the story and portrait treatment together so the subject, context and captions remain clear on a mobile screen.
- FOR
- Social producers, communications teams and video editors
- SOURCE
- A landscape interview with one or two visible speakers
- OUTPUT
- A portrait composition with reviewed crop safety
- TARGET
- 30 to 90 seconds
01 · The instruction
Copy this editing brief.
Replace any bracketed variables, select the relevant recordings and keep the review checkpoint before rendering.
EDIT BRIEF
Create a 60-second 9:16 edit from this landscape interview around [story or argument]. Compose the complete spoken narrative first, then select a supported portrait framing treatment that keeps the active subject visible. Use face-aware crop evidence where safe, retain a wider or stacked treatment when two people must remain visible and refuse any crop that cuts through a required face. Keep captions inside mobile-safe margins and away from important facial detail. Return the source-grounded cut plan and crop or QC information for review before rendering. Use Anjin Media directly, or the Anjin Media MCP when running this through an AI agent, so the proposed edit remains grounded in the selected source footage and can be reviewed before rendering.
02 · Source material
Give the brief enough evidence to work with.
A landscape source with sufficient resolution and usable subject framing.
The required speakers or objects that must remain visible.
A portrait-safe caption style and brand margin settings.
A fallback preference when a safe single crop is not possible.
03 · Representative example
Portrait framing follows the subject and admits its limits
A representative two-person interview uses a close crop while one contributor speaks and a supported wider treatment when both faces are necessary. Unsafe overlap is not presented as successful tracking.
SOURCE
1920 × 1080 interview
SUBJECTS
2 speakers
OUTPUT
1080 × 1920
QC
Face coverage reviewed
04 · Reviewable plan
The proposed cut structure.
OPEN
02:14–02:25
Speaker A crop
Keep the opening subject comfortably inside the portrait frame.
QUESTION
02:26–02:33
Supported two-person treatment
Retain both speakers where relationship matters.
ANSWER
02:34–03:02
Speaker B crop
Follow the active contributor through the explanation.
CONTEXT
03:03–03:16
Wide fallback
Avoid an unsafe crop during overlapping faces.
CLOSE
03:17–03:28
Speaker B crop
Return focus to the concluding speaker.
05 · Editorial reasoning
Why this composition works.
- 01The narrative determines which subjects must be visible.
- 02Single-subject crops use detected face position rather than a fixed centre.
- 03The treatment changes when two faces cannot fit safely.
- 04Caption margins are reviewed as part of composition, not added blindly afterward.
Review before render
Keep human judgement in the loop.
- Inspect face coverage across the full segment, not one sample frame.
- Check transitions between crop treatments.
- Test captions at mobile size.
- Confirm the fallback does not imply unsupported speaker-following behaviour.
06 · Change the angle
Alternative briefs to request.
A centre-safe single speaker
Create a 45-second portrait edit using only source moments where the active speaker can remain safely framed in 9:16. Exclude visually unsafe sections and return crop QC. Use Anjin Media directly, or the Anjin Media MCP when running this through an AI agent, so the proposed edit remains grounded in the selected source footage and can be reviewed before rendering.
A two-person stacked version
Create a 60-second two-person portrait edit using the supported stacked treatment where both speakers must remain visible. Preserve speaker order and caption-safe space. Use Anjin Media directly, or the Anjin Media MCP when running this through an AI agent, so the proposed edit remains grounded in the selected source footage and can be reviewed before rendering.
A square LinkedIn version
Create a 60-second 1:1 version of the approved narrative. Re-evaluate framing and captions for square output instead of scaling the portrait render. Use Anjin Media directly, or the Anjin Media MCP when running this through an AI agent, so the proposed edit remains grounded in the selected source footage and can be reviewed before rendering.
07 · Boundaries
Common failure points.
A fixed centre crop misses the speaker
Review detected subject position rather than assuming the centre of the landscape source is safe.
Two faces compete for one portrait window
Use a supported alternative or refuse the unsafe crop.
Captions cover the subject
Apply brand margins and inspect text density against the actual crop.
Different aspect ratios reuse one crop
Plan each aspect from the approved composition and source geometry.
Questions about landscape interview to vertical edit
- Is face-aware reframing facial recognition?
- No. Anjin uses detected face positions for framing. It does not identify a person's real-world identity from their face.
- Can Anjin keep two speakers in a portrait frame?
- Supported framing can include a stacked treatment. Unsafe multi-face crops should be refused or use an appropriate fallback.
- Does it support 9:16 output?
- Yes. Supported portrait output is 1080 by 1920.
- Can crop quality be reviewed?
- Available crop and QC information accompanies the workflow for review.
Use the instruction
Start with the footage you already have.
Copy the brief, select the relevant recordings and review the proposed structure before rendering.