Direct first, generate later: 3D blocking workflow for choreographed AI fight videos
HeyAmit_ · x · 2026-09-18
The author choreographed a full AI fight sequence using Pippit's 3D Director Studio: arranging moves on a timeline with the Action Library and directing the camera beat by beat — fast during action, slow and unsettling at mood shifts.
Pipeline: 3D blocking → action planning → camera direction → reference export → cinematic generation with Seedance 2.5. The final video kept the choreography, timing and story as planned. Key idea: direct the scene first instead of letting AI invent everything.
Related event: Choreographing AI Fight Scenes with Pippit's 3D Director Studio(2 posts)→
More from Multimodal
- Free course covers the 4 hard problems of production generative media agents — Al_Grigor · 2026-11-04
- PixelBench: testing multimodal models pixel-by-pixel instead of with QA — bdsqlsz · 2026-11-04
- Kitten-TTS 2 offers free voice cloning on laptops, no GPU needed — solyarisoftware · 2026-11-04
- "We'll Take Your RAM": a fully local-hardware cyberpunk meme from 2030 San Francisco — Wonderful-Yak-3348 · 2026-11-04
- Openfilm: open-source AI video agent turns coding agents into video makers — solyarisoftware · 2026-11-04
- Microsoft reopens VibeVoice: 10-second voice cloning, 90-minute local TTS — solyarisoftware · 2026-11-04