GaussiAnimate turns 4D captures into rigged 3D assets without physics simulation

janusch_patas · x · 2026-10-12

Researchers from Westlake University, HKU and collaborators present GaussiAnimate at SIGGRAPH Asia 2026: it compresses consistent 4DGS or mesh sequences into rigged, animatable 3D assets while preserving secondary dynamics (loose clothing, soft tissue) — no physics simulation or learned deformable model required.

Key ideas:

Code and the arXiv paper are available.

Original post →

More from Multimodal

Multimodal channel →