Why this structure is useful
A still image already makes many choices. The prompt should preserve those choices first, then request only the movement that the image can support.
Copyable template
Use the supplied reference image as the first-frame visual anchor. Preserve [subject identity], [composition], [lighting], and [palette]. Add only [primary motion] with [secondary environmental motion]. Camera: [camera behavior]. Keep the sequence in [format] for [duration]. Do not [specific changes to avoid].Variables to change
- subject identity and composition
- lighting and palette
- primary and environmental motion
- camera behavior
- changes to avoid
Three-shot plan
- Shot 1: Begin from the supplied composition without reframing the subject.
- Shot 2: Introduce the primary motion while keeping the subject readable.
- Shot 3: Let the motion settle into a frame that still matches the reference.
Failure modes to avoid
- Asking for a new location, outfit, or angle turns a reference-led task into a redesign.
- Large camera moves can expose missing background information and cause unwanted visual changes.
Continue planning
Open this template in the prompt builder
Review the Seedance 2.5 status tracker before relying on model access, pricing, or API claims.