Generalizable Articulated Object Perception with Superpoints

TL;DR

This paper presents a superpoint-based perception method that enhances 3D part segmentation of articulated objects, leveraging learnable superpoints, SAM, and transformer decoders for improved accuracy across categories.

Contribution

Introduces a novel superpoint generation technique combined with SAM and transformer decoders for superior 3D part segmentation of articulated objects.

Findings

Outperforms state-of-the-art in cross-category segmentation

Achieves 77.9% AP50 on seen categories

Achieves 39.3% AP50 on unseen categories

Abstract

Manipulating articulated objects with robotic arms is challenging due to the complex kinematic structure, which requires precise part segmentation for efficient manipulation. In this work, we introduce a novel superpoint-based perception method designed to improve part segmentation in 3D point clouds of articulated objects. We propose a learnable, part-aware superpoint generation technique that efficiently groups points based on their geometric and semantic similarities, resulting in clearer part boundaries. Furthermore, by leveraging the segmentation capabilities of the 2D foundation model SAM, we identify the centers of pixel regions and select corresponding superpoints as candidate query points. Integrating a query-based transformer decoder further enhances our method's ability to achieve precise part segmentation. Experimental results on the GAPartNet dataset show that our method outperforms existing state-of-the-art approaches in cross-category part segmentation, achieving AP50 scores of 77.9% for seen categories (4.4% improvement) and $39.3\%$ for unseen categories (11.6% improvement), with superior results in 5 out of 9 part categories for seen objects and outperforming all previous methods across all part categories for unseen objects.