Efficient Representations of Object Geometry for Reinforcement Learning of Interactive Grasping Policies

Open in new window