Preprint
Aug 2026
UniMoFlow: Grounding Instruction-Driven 3D Human Motion Editing in Generation
A closed-loop synthesis-and-verification pipeline that produces Omni-MoEdit, a large-scale dataset spanning body-part, amplitude, temporal, action, and style edits, and a unified latent flow-matching model that shares broad semantic and kinematic knowledge between generation and editing is developed.
Yilei Hua, Beibei Jing, Ce Zheng et al.
· 0 citations