Reference to Video
Upload one or more reference images to guide style, subject details, and scene direction. The model uses them as creative guidance rather than fixed frames.
How to use this mode
- • Reference images shape visual style and content direction instead of locking any single frame.
- • This page supports models such as Wan 2.7, Seedance 2.0, Seedance 2.0 Fast, and selected Doubao Seedance variants.
- • Combining multiple references is useful when you need stronger style control, worldbuilding, or character consistency.
- • Pair the references with a precise prompt to improve alignment between the uploaded images and the generated motion.
Content policy notice
Any non-compliant generation will fail — this is not a scam or a site error. The Seedance 2.0 model has recently tightened content restrictions, and all real human face inputs (selfies, portraits, celebrities) are now blocked at the model level. Violent and NSFW content is also rejected. You can still create amazing videos with illustrations, anime characters, AI-generated faces, or stylized artwork.
Best use cases and prompt ideas
Style transfer
Upload a set of images with a defined visual language and generate video that follows the same tone, palette, and composition style.
Concept blending
Combine visual elements from several images to create a new scene that inherits details from each reference.
Scene building
Provide references for environment, characters, props, and wardrobe when the final shot needs more structured direction.
Art direction
Use illustrations, keyframes, or campaign stills as references when you want the output to stay close to an approved look.
- • Single reference: Use this image’s color palette and framing to create a calm sunset shoreline sequence.
- • Multiple references: Blend these images into a dreamlike forest scene with drifting light particles and butterflies.
- • Stylized output: Follow the illustration style of these references and generate a cartoon city street with pedestrians crossing the frame.