ZeroBP: Learning Position-Aware Correspondence for Zero-shot 6D Pose Estimation in Bin-Picking

TL;DR

ZeroBP introduces a zero-shot 6D pose estimation framework for bin-picking that learns position-aware correspondences, effectively handling textureless and ambiguous workpieces, and outperforms existing methods on the ROBI dataset.

Contribution

The paper presents ZeroBP, a novel zero-shot pose estimation method that leverages position-aware correspondences to improve accuracy in challenging bin-picking scenarios.

Findings

ZeroBP achieves 9.1% higher average recall than state-of-the-art methods.

ZeroBP effectively handles textureless and ambiguous workpieces.

Extensive experiments validate the superiority of ZeroBP on the ROBI dataset.

Abstract

Bin-picking is a practical and challenging robotic manipulation task, where accurate 6D pose estimation plays a pivotal role. The workpieces in bin-picking are typically textureless and randomly stacked in a bin, which poses a significant challenge to 6D pose estimation. Existing solutions are typically learning-based methods, which require object-specific training. Their efficiency of practical deployment for novel workpieces is highly limited by data collection and model retraining. Zero-shot 6D pose estimation is a potential approach to address the issue of deployment efficiency. Nevertheless, existing zero-shot 6D pose estimation methods are designed to leverage feature matching to establish point-to-point correspondences for pose estimation, which is less effective for workpieces with textureless appearances and ambiguous local regions. In this paper, we propose ZeroBP, a zero-shot pose estimation framework designed specifically for the bin-picking task. ZeroBP learns Position-Aware Correspondence (PAC) between the scene instance and its CAD model, leveraging both local features and global positions to resolve the mismatch issue caused by ambiguous regions with similar shapes and appearances. Extensive experiments on the ROBI dataset demonstrate that ZeroBP outperforms state-of-the-art zero-shot pose estimation methods, achieving an improvement of 9.1% in average recall of correct poses.