None
None
Long Story Video Skill
Turn your idea into 10s–10min video—AI writes scripts, prompts, and generates footage automatically.
Ads Video Skill
Generate professional ads and sales videos—AI auto-generates scripts, prompts, and footage.
3D Science Explainer Video Skill
Convert scientific concepts into stunning 3D explain animations
Feedback
freeTrialImage.bannerPity
freeTrialImage.upgradeUnlock
- ✓freeTrialImage.benefitHd
- ✓freeTrialImage.benefitWatermark
- ✓freeTrialImage.benefitUnlimited
Minimax H3 Video To Video
Give an edited video a new lineup of performers. Minimax H3 Video To Video changes only who appears on screen — pacing, lens work and audio stay as shot.
All Tools
Discover our comprehensive AI-powered animation toolkit
MiniMax H3
MiniMax H3 AI Video Generator
Seedance 2.5
The Future of AI Video Is Here.
Seedance 2.0
The Future of AI Video Is Here.
Veo3.1
Create Stunning Videos with Veo3.1
FLUX 3 Video Generator
Kling 3.0
Next-Gen AI Video Generator
Grok Video Generator
Create Videos from Text or Images with AI
MiniMax H3 video generator
MiniMax H3 AI Video Generator
What Minimax H3 Video To Video Can Do for Your Footage
Think of it as re-casting an existing scene rather than shooting a new one — the original performance, lighting and rhythm are kept, and only the faces are replaced.
- Casting replacement, not generationNothing here is rendered from scratch: the tool works on top of footage you have already captured, so the original visual character is never rebuilt.
- Up to four performers, cast from photosSupply a single clear portrait per newcomer — four at most — and they inherit the main on-screen parts, assigned in left-to-right order in every shot those parts appear.
- Your footage keeps its performanceGestures, timing, illumination, lens movement, cutting rhythm and the original soundtrack all remain, so the output drops straight into your locked edit.
Minimax H3 Recast in Three Steps
The workflow takes three moves — choose a clip you are happy with, organise one portrait per newcomer, then render the replacement.
Where Minimax H3 Recast Excels
The performance, the cut and the audio of your original footage all survive — only the people appearing in it are replaced.
Four faces, ordered left to right
Four portraits is the ceiling; they claim the principal parts in left-to-right sequence in every scene those parts appear. Identity tracking continues across cuts, and a short prompt can change the assignment.
Same performance, different people
Everything about the acting is inherited from your source — timing, expression, movement, lighting — while the new faces are drawn from the images you provide.
Editing rhythm and audio preserved
Your cutting rhythm and original soundtrack pass through unchanged, which means the output can go directly into a finished timeline with no re-editing.
Affordable 768p checks, 1080p masters
Verify the result with a cheaper 768p pass before paying for the higher resolution; since the output mirrors your input duration, cost scales with clip length.
Hair and wardrobe come from your photos
If your reference image captures more than the face, hairstyle and garments transfer as well — a half-body photo therefore outperforms a close-up headshot.
Made for footage that already works
Production teams, brand marketers, social creators and fashion or commerce studios reach for it whenever a take is flawless but features the wrong person.
Minimax H3 Video To Video: Questions Answered
Answers to what people ask most about replacing performers in video they have already shot.
What exactly does Minimax H3 Video To Video do?
It replaces the people visible in an existing video with the faces from images you upload. The original motion, camera movement, editing and sound are left in place, and unless you say otherwise, your photos fill the main roles in left-to-right order in every scene those roles appear in.
Is there a length limit on the source video?
Between 5 and 30 seconds, with no single shot running past 15 seconds, so cut longer material into pieces first. The result is delivered at exactly the same duration as your source.
How many performers can be replaced, and in which order?
Up to four, with one image per person, and by default they occupy the principal roles in left-to-right sequence. You can override that mapping with a short instruction, for example naming which person on screen becomes which reference.
Which reference photos give the best results?
Choose one crisp, well-lit, front-facing image per person. A half-body frame transfers hair and garments more faithfully than a close headshot. Steer clear of sunglasses, hats, hands covering the face, group shots, and blurry or heavily filtered pictures.
Do I need 768p or 1080p?
Use 768p first to make sure every face looks right, then produce the 1080p version for delivery. Cost is driven mainly by clip duration rather than by resolution on its own.
Are there cases where it will not work well?
Avoid it if you need fresh action, different camera moves or a new setting. Quality also drops with crowded scenes, tiny or far-away faces, rapid movement, or subjects hidden after a hard cut — and in long takes or profile angles, the likeness may drift back toward the original performer.
Recast a Finished Clip with Minimax H3 Video To Video
Drop fresh faces into a video that is already cut and graded, and keep the pacing, camera work, edits and soundtrack exactly where they were. Afterwards, explore the Recast guide or weigh up the related MiniMax, Hailuo and Kling models.