DigestOps
Archived signal · AI Web & Product Design
Zentrix⌚️@ZentrixHQx

Consistent cinematic video series from one ChatGPT Image 2 character

  • Create a character sheet with front, side, back views, 3-5 facial expressions, and different lighting conditions as source of truth.
  • Generate reference scenes (wide, medium, close-up) keeping clothing, hairstyle, and accessories identical.
  • Use cinematographer-style prompts like 'Low-angle tracking shot, 35mm lens, cinematic lighting, shallow depth of field' instead of generic descriptions.
how to reproduce · 1. Generate character sheet: front, side, back views, expressions, lighting. 2. Generate reference scenes: wide, medium, close-up in different environments. 3. Prompt with cinematographer language. 4. Build every keyframe before animating.
909.4K views↗ Open the original
Share
Where this lives now

This signal is part of the DigestOps evidence archive. The living answers built on evidence like this are at AI Web & Product Design; agents can query it as JSON at /api/v1/cards/83 or over MCP.