Jul 2026
4DHumanDiff: Direct Text-to-4DGS Generation for Consistent 360-Degree Dynamic Humans
4DHumanDiff is presented, a diffusion framework that directly generates dynamic humans represented by 4D Gaussian Splatting from text prompts and achieves better temporal and multi-view consistency, and reduces inference time by more than 10x.
Ren-Rong Wu, Haoran Chen, Yuxiang Wei et al.
· arXiv.org · 0 citations