3D Diffuser Actor: Policy Diffusion with 3D Scene Representations

arXiv:2402.10885v1 Announce Type: cross
Abstract: We marry diffusion policies and 3D scene representations for robot manipulation. Diffusion policies learn the action distribution conditioned on the robot and environment state using conditional diffusion models. They have recently shown to outperform both deterministic and alternative state-conditioned action distribution learning methods. 3D robot policies use 3D scene feature representations aggregated from a single or multiple camera views using sensed depth. They have shown to generalize better than their 2D counterparts across camera viewpoints. …

abstract actor arxiv cs.ai cs.cv cs.lg cs.ro diffusion diffusion models distribution environment learn manipulation policy robot robot manipulation state type