Hand-Object Interaction Controller (HOIC): Deep Reinforcement Learning for Reconstructing Interactions with Physics

TL;DR

This paper introduces HOIC, a deep reinforcement learning approach that reconstructs realistic hand-object interactions from RGBD data by incorporating physics-based control and contact models, improving accuracy without heuristic rules.

Contribution

The work presents a novel deep reinforcement learning method with object compensation control and physics-based contact modeling for accurate hand-object interaction reconstruction.

Findings

Successful reconstruction of complex hand-object interactions.

Improved physical plausibility without heuristic rules.

Enhanced stability through object compensation control.

Abstract

Hand manipulating objects is an important interaction motion in our daily activities. We faithfully reconstruct this motion with a single RGBD camera by a novel deep reinforcement learning method to leverage physics. Firstly, we propose object compensation control which establishes direct object control to make the network training more stable. Meanwhile, by leveraging the compensation force and torque, we seamlessly upgrade the simple point contact model to a more physical-plausible surface contact model, further improving the reconstruction accuracy and physical correctness. Experiments indicate that without involving any heuristic physical rules, this work still successfully involves physics in the reconstruction of hand-object interactions which are complex motions hard to imitate with deep reinforcement learning. Our code and data are available at https://github.com/hu-hy17/HOIC.