Contact2Grasp: 3D Grasp Synthesis via Hand-Object Contact Constraint

TL;DR

Contact2Grasp introduces a two-stage approach for 3D grasp synthesis that leverages contact constraints to improve the efficiency and generality of grasp pose generation, validated by extensive experiments.

Contribution

The paper proposes a novel contact-constrained framework that factorizes grasp synthesis into contact map generation and pose prediction, enhancing performance over existing methods.

Findings

Outperforms state-of-the-art methods on multiple metrics

Effective use of contact constraints improves grasp generation quality

Penetration-aware optimization refines grasp poses

Abstract

3D grasp synthesis generates grasping poses given an input object. Existing works tackle the problem by learning a direct mapping from objects to the distributions of grasping poses. However, because the physical contact is sensitive to small changes in pose, the high-nonlinear mapping between 3D object representation to valid poses is considerably non-smooth, leading to poor generation efficiency and restricted generality. To tackle the challenge, we introduce an intermediate variable for grasp contact areas to constrain the grasp generation; in other words, we factorize the mapping into two sequential stages by assuming that grasping poses are fully constrained given contact maps: 1) we first learn contact map distributions to generate the potential contact maps for grasps; 2) then learn a mapping from the contact maps to the grasping poses. Further, we propose a penetration-aware optimization with the generated contacts as a consistency constraint for grasp refinement. Extensive validations on two public datasets show that our method outperforms state-of-the-art methods regarding grasp generation on various metrics.