Hand-Object Contact Detection using Grasp Quality Metrics

TL;DR

This paper introduces a new hand-object contact detection method leveraging grasp quality metrics from hand and object poses, achieving high accuracy and aiming for real-time vision-based applications in robotic handover systems.

Contribution

The paper presents a novel contact detection approach based on grasp quality metrics, with high accuracy demonstrated on the DexYCB dataset, and plans for real-time vision integration.

Findings

Achieved nearly 90% accuracy in contact detection.

Validated the approach on the DexYCB dataset.

Future work includes real-time vision-based implementation.

Abstract

We propose a novel hand-object contact detection system based on grasp quality metrics extracted from object and hand poses, and evaluated its performance using the DexYCB dataset. Our evaluation demonstrated the system's high accuracy (approaching 90%). Future work will focus on a real-time implementation using vision-based estimation, and integrating it to a robot-to-human handover system.