Add motion to copy
Video duration: 3-30 seconds
Add your character
Image with visible face and body
When Character Orientation matches the video, complex motions perform better; when it matches the image, camera movements are better supported.
Loading library...
Super Promotion
Create stunning AI photos & videos with essential tools
Unlock the Basic Plan for just $1
Auto-renewal is active. Cancel anytime. 90% off applies to the first billing cycle.
By choosing your age and continuing you agree to our Terms of Use and Privacy Policy
Please review before continuing
Kling Pro Motion Control Video Editing Guide
Kling Pro Motion Control is available in Cleep for video workflows where movement is guided by source material rather than described only with text. A motion-control task involves at least two responsibilities: choosing a visual subject that can be animated clearly and choosing motion reference material whose timing and framing fit that subject. Careful preparation matters more than adding a long prompt.
Prepare the visual source
Use a sharp image or supported source with the complete subject visible. Limbs, clothing edges, and the surrounding floor or background should be easy to distinguish. Match the target aspect ratio before generation and leave room for the movement. Cropped hands, feet, or props can create ambiguity when the referenced motion requires them later.
Keep an untouched original and verify that you have permission to animate the person, character, product, or artwork. If identity must remain stable, avoid sources with heavy filters, extreme perspective, or very small faces.
Choose and align the motion reference
Select a reference with one clearly visible performer or movement, stable timing, and framing that is compatible with the visual source. Complicated occlusion, rapid cuts, large differences in body proportions, or movement outside the frame can make the transfer harder to review. Trim the reference to the useful action and decide which part of the movement is essential.
Prompt for context, not conflicting motion
Let the motion reference guide the action. Use the text instruction for environment, camera treatment, lighting, pace, and details that are not already defined. Avoid describing a second, conflicting movement. If the result needs a different action, change the reference or isolate that action in a separate shot.
Step-by-step workflow
- Prepare the subject source at the intended crop and quality.
- Trim the motion reference to one coherent action.
- Check that framing, direction, and visible body area are compatible.
- Add a concise prompt for scene and camera context.
- Generate a short test and review identity and motion before increasing complexity.
Quality review
Watch the complete output for identity drift, anatomy changes, sliding feet, unstable clothing, changing props, background distortion, and contact errors. Compare the rhythm and direction with the reference, but also judge whether the result looks physically coherent for the new subject. Inspect any transition where a limb or prop crosses the body.
Store the source, reference, prompt, settings, and output together. For a sequence of shots, approve them individually and maintain a production note for consistent framing, wardrobe, lighting, and disclosure.
Responsible use
Do not transfer motion to a real person’s likeness without permission or use an edit to fabricate an endorsement or event. Respect rights in both the visual source and the motion reference. Label synthetic footage when the context or publishing platform requires it.
Frequently asked questions
Why does the result differ from the reference motion?
Framing, proportions, occlusion, source quality, and incompatible camera direction can affect the transfer. Simplify the reference and align the inputs more closely.
Should the prompt repeat the action?
Usually the reference should define movement. Use text for setting and visual treatment unless a short clarification is needed.
What is the best first test?
Use a short, clearly framed movement with a sharp source and minimal camera complexity, then review identity and contact points before refining.