Studio product
Interface preview
Use an image as the visual starting point for your video.
Your image is uploaded before the video task is submitted.
Interface preview
Interface preview
Interface preview
Interface preview
Animate an approved source image while directing subject motion, camera behavior, scene continuity, duration, resolution, and optional native audio.
Choose a clear frame that establishes the subject, composition, product shape, wardrobe, or visual style.
Identify the visual details that should stay stable while the shot moves and develops.
Direct subject action, camera movement, environmental changes, pace, sound, and the intended final frame.
Review identity, geometry, textures, screen direction, camera movement, and audio throughout the full clip.
Turn an approved product still into a hero shot with deliberate camera movement and lighting.
Animate a portrait or character frame while preserving recognizable identity and art direction.
Add depth, atmosphere, environmental movement, and cinematic pacing to a finished image.
Image to video inherits composition and appearance from a still frame, then adds time. The best source already contains the identity, product geometry, viewpoint, and visual hierarchy that must survive motion; the prompt concentrates on change.
Use a sharp image with readable edges, unobscured hands, consistent reflections, and enough space for the intended camera move. Check labels, faces, accessories, and background geometry before uploading. A distorted or heavily compressed source can turn a minor flaw into a moving continuity problem.
List what must remain fixed—face identity, clothing, product shape, logo orientation, camera side, or background layout. Then name subject motion, environmental motion, and camera motion separately. This prevents a request for gentle animation from accidentally redesigning the object or changing the entire composition.
Infer a plausible next action from the pose and spatial layout. A seated subject can turn, speak, or reach; a product can rotate while highlights travel across its surface. Sudden changes of viewpoint or anatomy demand information the source does not contain and should be tested as separate shots rather than hidden inside one prompt.
Watch for facial drift, rubber-like geometry, changing label text, sliding contact points, unstable shadows, and backgrounds that breathe or bend. Pause at the first frame, midpoint, and ending, then compare them with the source. Revise the smallest failing category while preserving motion and camera instructions that already work.
Image-to-video generation uses an uploaded still as the visual starting point and a text prompt to direct motion and sound.
Use a sharp, clean image with a readable subject and composition that already contains the details the video must retain.
Explain what moves, how the camera behaves, what changes in the environment, and how the sequence should end.
Yes. The website uploads the selected image first, then sends its returned URL with the video-generation request.
Check the required input, choose output settings, and review the workflow-specific guidance above before starting.
Review the controls →