Fixing Dance Accuracy in Seedance 2.0 Using Depth Maps
jordiponsdotme · x · 2026-09-01
Addresses the choreography drift issue in Seedance 2.0 by introducing a depth map workflow.
The Core Problem: Feeding raw video forces the model to interpret variables like the person, clothes, and lighting, which confuses the motion capture.
The Solution: Convert source dance footage into a depth map, stripping away appearance and environment, leaving only the motion data.
The Workflow:
- Lock Character: Use GPT Image 2 to lock in face, build, and costume first.
- Extract Depth: Convert source dance footage into a depth map instead of using raw video.
- Feed Model: Input the depth map as the motion reference.
More from Multimodal
- Luma shares filmmaking-style workflow: motion-capture real stunts then restyle with video-to-video — lmoroney · 2026-10-18
- MOSS-Image-SFT at 4 Steps Draws Hype: 320x320 to 2K Without Artifacts — Powerful_Evening5495 · 2026-10-18
- MOSS SFT vs Flux 3 Klein vs SD 3.0: Same-Prompt Comparison Exposes SD3's Age — ZootAllures9111 · 2026-10-18
- Dev builds Mumbai-inspired 3D manga background generator from OpenStreetMap data — CurieuxExplorer · 2026-10-18
- Omni 1'1 flash video demo generates atmospheric foggy Victorian haunted house — Dense-Instruction210 · 2026-10-18
- Flux 3 klein distilled tested: int8 workflow delivers surprisingly good 2MP images — Ok_Tale7582 · 2026-10-18