Edge Grasp Network: A Graph-Based SE(3)-invariant Approach to Grasp Detection

TL;DR

This paper introduces a graph-based neural network model that detects 6-DoF grasp poses from point clouds, improving grasp success rates and handling single-view data from arbitrary angles.

Contribution

It presents a novel SE(3)-invariant approach using graph neural networks for grasp detection from point clouds, enhancing robustness and accuracy.

Findings

Achieves higher grasp success rates than existing methods

Works effectively with single-view point clouds from any viewing angle

Demonstrates improved robustness in grasp detection

Abstract

Given point cloud input, the problem of 6-DoF grasp pose detection is to identify a set of hand poses in SE(3) from which an object can be successfully grasped. This important problem has many practical applications. Here we propose a novel method and neural network model that enables better grasp success rates relative to what is available in the literature. The method takes standard point cloud data as input and works well with single-view point clouds observed from arbitrary viewing directions.