Motion control takes two inputs: a clip that carries the movement, and an image of whoever should be doing it. The clip is the driver. Length, beat, gesture path and (sometimes) camera come from that file. Wardrobe, face, set and product identity come from the still and the prompt. You are casting a performance, not describing one. 'She spins twice and points at the label' is a paragraph with an uncertain outcome; two seconds of someone doing it is the brief.
It is not A-roll unless you put it in the cut. It is not reference-to-video. A reference image is an identity the model may place anywhere; a driving clip is a timeline of poses. Crowded frames, extreme close-ups and heavy occlusion make a bad driver because the pose the model extracts is only as good as what it could see. Full-body, clean background, readable limbs transfer. A talking-head close-up will not give you a dance.
Kling Video V3 Pro Motion Control is the catalog row that names this job in plain language: still plus driving clip. Facial performance usually does not ride along; lipsync is a later pass if the mouth has to match a track. Do not expect to edit the driver after the generate — the output is a new take wearing that motion, not a layer you can retime independently.
In practice
- Shoot or pick the driver for readable pose: whole body, uncluttered background, one action.
- Match length to the driver. Padding the still with a longer prompt will not invent beats that were not in the clip.
- Keep the driver out of the finished film unless you meant to show that performer.
Motion-control models
Catalog entries that transfer movement from one clip onto another subject. 7 of the 351 models in the Versely catalog qualify.
| Model | Provider | Type |
|---|---|---|
| Kling Video V3 Pro Motion Control | Kling | Video |
| Kling Video V3 Standard Motion Control | Kling | Video |
| DreamActor V2 | ByteDance | Video |
| Wan 2.2 14B Move | Wan | Video |
The mistake to avoid
Feeding a messy phone clip and blaming the model when limbs teleport. The driver is the motion spec; garbage pose in is garbage pose out.
Where you will run into it
- AI Video Generator — Text-to-video, image-to-video, and story chaining in one studio.
- Kling Video V3 Pro Motion Control — Kling V3 Pro motion control - transfer motion from a driving video to a reference image at pro-tier quality
Related terms
Motion control
AI motion control copies the movement and expressions from a reference video onto a character image, as in Kling Motion Control. The performer changes.
Reference-to-video
Reference-to-video means the model builds a clip from separate images of a subject it can place anywhere. Those images are an identity, not the first frame.
A-roll
A-roll is the primary footage a cut is built around — usually a person speaking — as distinct from B-roll that covers it and from a generate that tries to be both.
Text-to-video
Text-to-video is generation from a written prompt alone — you describe a shot, the model invents every frame of it, and no image or footage goes in.
Image-to-video
Image-to-video animates a still you supply: the picture becomes the opening frame, and the prompt describes only what happens next.
The all-in-one AI studio for creators. 60+ models for video, image, voice, music and lipsync, in your browser or on your phone.
Free on the web and on iPhone, one account.