D-Grasp: Physically Plausible Dynamic Grasp Synthesis for Hand-Object Interactions
This addresses the challenge of dynamic hand-object interactions for robotics or animation, representing an incremental advance in grasp synthesis.
The paper tackles the problem of generating physically plausible hand motions to grasp and move objects to target poses, proposing a reinforcement learning method that achieves stable grasps and a wide range of motions.
We introduce the dynamic grasp synthesis task: given an object with a known 6D pose and a grasp reference, our goal is to generate motions that move the object to a target 6D pose. This is challenging, because it requires reasoning about the complex articulation of the human hand and the intricate physical interaction with the object. We propose a novel method that frames this problem in the reinforcement learning framework and leverages a physics simulation, both to learn and to evaluate such dynamic interactions. A hierarchical approach decomposes the task into low-level grasping and high-level motion synthesis. It can be used to generate novel hand sequences that approach, grasp, and move an object to a desired location, while retaining human-likeness. We show that our approach leads to stable grasps and generates a wide range of motions. Furthermore, even imperfect labels can be corrected by our method to generate dynamic interaction sequences.