HumanSkill-4D: training-free language queries can animate Gaussian avatars, with spatial and temporal grounding
HumanSkill-4D: Training-Free Language Querying of Animatable Gaussian Avatars in Space and Time
HumanSkill-4D proposes a training-free framework that uses a shared multimodal large language model and geometric operators to select local regions of an animatable Gaussian avatar from language queries, and can specify a time period matching a pose or action. The method binds body parts and clothing categories to persistent Gaussian regions via semantic memory, and composes operator programs using driving joint trajectories in motion memory, making selections traceable. The team built three human-annotated benchmarks, on which it outperforms baselines in part segmentation and 3D and 4D referring segmentation, with only a few minutes of setup per avatar.
Open original arxiv.org