A 3D blocking pass gives AI video frame-level control before Seedance 2.0 renders
andrew_n_carr · x · 2026-07-24
The post lays out a 3D-first workflow for directing AI video instead of relying on pure prompt lottery.
The team writes the character and scene first, builds four-view reference sheets, creates a 3D blocking pass in Maya/Blender/Unreal, and then feeds the motion reference plus visual references into Seedance 2.0 for neural rendering. The goal is frame-by-frame control over timing, camera, and spatial relationships before the video model does the final render.
More from Multimodal
- ComfyUI Krea2 Prompt Styler Node Revived in a Fork With a Massive Style Library — Winter_unmuted · 2026-10-27
- Single HTML file generates retro poster designs in-browser with zero LLM calls — Afinetheorem · 2026-10-27
- A Redditor Used Copilot to Imagine and Archive Their Own '80s/'90s Winter Streetwear Concepts — Valuable-Sky2134 · 2026-10-27
- Transkun open-source piano transcription model nails MIDI, sustain events included — alexanderchen · 2026-10-27
- fal adds FLUX 3 Image Klein LoRA training: 10 reference images, 3 LoRAs per request — OdinLovis · 2026-10-27
- Guy turns his girlfriend's selfie into a playable AI game ad about popping pimples — CurieuxExplorer · 2026-10-27