What this page is for
Use this workflow when you have a reference image, product frame, ad still, or storyboard image and want to create AI video direction for Kling, Veo, Sora, Runway, or similar image-to-video models.
Reference frame to AI video direction
Image to video prompt
Convert a still image into video prompt language with camera movement, subject action, scene continuity, atmosphere, and commercial pacing.
What the prompt should capture
A useful video prompt should preserve the subject and scene while adding camera movement, motion timing, atmosphere, continuity, and constraints against warped objects, text artifacts, or unstable subject identity.
How to use it
Open the homepage tool, upload a frame, choose Product Shot or Marketing Visual mode, then copy Video Motion. Use it as the motion layer and combine it with model-specific settings in your video tool.
Prompt examples
These are example structures. Use the homepage tool to generate a prompt from your own reference image.
Product reveal motion
slow dolly-in toward hero product, soft studio highlights moving across surface, clean commercial background, premium detail, stable object shape, 4 second product reveal, no text artifacts
Lifestyle ad motion
gentle handheld push-in, subject interacts naturally with product, warm room light, subtle parallax, authentic UGC feel, commercial pacing, stable face and hands
Cinematic brand frame
slow cinematic pan across scene, atmospheric light, smooth subject motion, preserve original composition and palette, high-end campaign film still, clean continuity, no warped geometry
FAQ
Does this generate the video?
No. PromptLens generates the text direction. Paste the Video Motion output into Kling, Veo, Sora, Runway, or another AI video tool.
Can I use product photos for video prompts?
Yes. Product frames work well because the model has a clear subject, lighting direction, and background to preserve during motion.
Which images work best?
Use clear frames with one main subject, visible lighting, and enough background context for camera movement. Busy collages and tiny screenshots usually produce weaker video prompts.