ContactGen: Generative Contact Modeling for Grasp Generation

TL;DR

ContactGen introduces a new object-centric contact representation for hand-object interaction, enabling the generation of diverse, high-fidelity grasps through a conditional generative model and optimization.

Contribution

The paper proposes ContactGen, a novel contact representation and generative framework for grasp synthesis, improving diversity and geometric feasibility of generated grasps.

Findings

Generates diverse, high-quality human grasps for various objects.

Uses a contact map, part map, and direction map for detailed contact modeling.

Achieves state-of-the-art results in grasp generation quality.

Abstract

This paper presents a novel object-centric contact representation ContactGen for hand-object interaction. The ContactGen comprises three components: a contact map indicates the contact location, a part map represents the contact hand part, and a direction map tells the contact direction within each part. Given an input object, we propose a conditional generative model to predict ContactGen and adopt model-based optimization to predict diverse and geometrically feasible grasps. Experimental results demonstrate our method can generate high-fidelity and diverse human grasps for various objects. Project page: https://stevenlsw.github.io/contactgen/