Browse AI Video Generation SECTION 7: Image-to-Video & Reference Workflows

The Hybrid Workflow Designer

3 fill-in slots · from The AI Prompt Handbook for AI Video Generation

Your details

0/3 filled

The Hybrid Workflow Designer

Design an image-to-video workflow for this project.

Project: [DESCRIBE]
Shots: [NUMBER AND DESCRIPTION]
Image tool: [NAME]
Video tool: [NAME]

Produce:
1. Which shots should start from a still and which should be
   text-to-video
2. For the image-to-video shots: what the still must contain
3. The image prompts for each
4. How to select the best still — what makes a good starting frame
   for video
5. The motion prompts to pair with each still
6. What to check before committing credits to the video generation

Principles to apply:
- Image generation is cheap, fast, and controllable; video
  generation is not. Push as much control as possible into the
  still.
- A still with a clear subject, clean composition, and space for
  the intended motion works best
- A still that is already visually busy leaves the model nowhere to
  put motion

Then: which shots in my project would improve most from this
approach, and which are genuinely better as text-to-video.

Highlighted [slots] are still empty. Downloads are plain .txt — paste into any AI assistant.