Product photo animation
Keep the product fixed, add a slow push-in, and constrain label, packaging, and reflections.
I2V prompt
Wan AI Image to Video Prompt Recipe
Use this image to video recipe to build a Wan AI AI video prompt with subject, action, camera, lighting, style, and constraints.
Copy-ready prompt
Create a coherent 5 seconds AI video in 16:9: the main subject from the reference image coming to life with subtle natural motion in the same environment shown in the reference frame. Use gentle push-in while preserving the original composition, matching the reference image lighting and color temperature, and image-to-video prompt, preserve identity, texture, and scene layout. Maintain subject consistency, stable scene continuity, and believable motion; avoid identity drift, changed outfit, altered background, warped face, extra objects.
Advertisement
Prompt notes
- This version uses a structured sentence that keeps the subject and scene coherent.
- subject: the main subject from the reference image
- action: coming to life with subtle natural motion
- setting: the same environment shown in the reference frame
- camera: gentle push-in while preserving the original composition
- lighting: matching the reference image lighting and color temperature
- style: image-to-video prompt, preserve identity, texture, and scene layout
- negative: identity drift, changed outfit, altered background, warped face, extra objects
- aspectRatio: 16:9
- duration: 5 seconds
- Asking for a large pose change can break the likeness from the reference image.
- Forgetting background constraints can make the scene quietly change across frames.
Use this when
- You have a still image, product shot, character frame, or concept art and need a short motion draft.
- The source image should stay recognizable while camera movement or environmental motion adds life.
- You want the prompt to protect identity, outfit, layout, lighting, and background consistency.
How to adapt each field
subject
Refer to the image subject directly and describe only the traits that must stay unchanged.
action
Use small believable motion first, such as breathing, hair movement, product rotation, or light movement.
setting
Tell the model to preserve the original environment unless you intentionally want a changed scene.
camera
Gentle motion usually works better than aggressive camera paths when an uploaded image is the anchor.
lighting
Match the reference lighting before adding new cinematic effects.
style
Use preservation language for identity, materials, composition, and scene layout.
negative
Call out identity drift, outfit changes, background changes, face warping, and added objects.
Drafting workflow
- Choose the parts of the image that must remain unchanged.
- Add one small subject motion or one small camera move, not both at full strength.
- Check whether identity, clothing, object shape, and background remain stable.
- Increase motion only after the reference-frame likeness is preserved.
Useful variations
Portrait motion
Use subtle blinking, breathing, and hair movement while protecting face shape and outfit.
Concept art reveal
Add atmospheric motion such as fog, light rays, or particles while preserving composition.
Image to Video prompt FAQ
Why should image-to-video prompts use preservation language?
The uploaded image already defines the subject and composition. Preservation language helps keep the generation anchored to that reference instead of reinventing the scene.
How much motion should I request from one image?
Start with subtle motion. Strong camera moves, pose changes, and background changes can make the output drift away from the source image.
Should the prompt repeat everything visible in the image?
No. Describe the important things that must stay stable, then let the reference image carry the rest of the visual detail.
Related prompt pages
Independent prompt drafting aid. Verify final prompts inside the current model interface.