Choose a readable driving video
Use one visible performer with limited occlusion, stable cuts, and motion that fits within the supported duration.
Motion transfer workspace
Make the character follow the performance.
A specialist motion-transfer workflow that maps movement from a driving video onto a reference character image.
Character image
Driving video
Motion result
Cargando...
Driving library
Check whether the motion is transferable
These clips are motion inputs, not generated outputs.
Fortune in Motion — driving reference
Example driving-video input from the motion library, not a generated output.
Martial arts — driving reference
A readable full-body motion input for testing performance transfer.
Gesture dance — driving reference
Example gesture sequence used as motion guidance, not model output proof.
Prepare the character image
Use a clear full- or half-body reference whose pose and framing make the target movement physically plausible.
Set orientation correctly
Match the character orientation to the image or driving video according to the input framing.
Review motion and identity
Check face stability, hands, feet, limb paths, clothing response, contact, and background artifacts frame by frame.
Motion-transfer prompt starters
Prompt 01
Dance recreation
Transfer the driving video's dance timing and full-body movement to the reference character. Preserve face, hair, outfit, and body proportions. Match orientation: [same as video/image].
Prompt 02
Sports action
Use the driving video as the exact motion guide for [sports action]. Keep the reference character recognizable, preserve limb timing, stable background, realistic weight and contact.
Prompt 03
Fashion performance
Transfer the reference performance to the supplied fashion character. Preserve garment shape and texture, natural fabric response, confident pacing, stable camera and body proportions.
Prompt 04
Gesture-led ad
Recreate the driving video's gesture sequence with the reference spokesperson. Keep face and outfit consistent; leave clear composition space for separately added campaign text.
Review before publishing
- This route requires reference media and is not a general text-only Kling generator.
- Occlusion, fast spins, floor contact, hands, and crossed limbs can introduce artifacts.
- Only use driving performances and character images you have permission to use.
Kling V3 Motion Control FAQ
What inputs does Kling V3 Motion Control need?
The connected workflow uses one character reference image and one driving video. The driving clip can be up to 30 seconds according to the current online configuration.
Is this text to video?
No. Prompt text can clarify the result, but the core task is transferring motion from a reference video to a reference character image.
How do I get cleaner motion transfer?
Use a single clearly visible performer, avoid heavy occlusion and rapid cuts, match body framing, and select the correct character orientation.