Generalizable and Actionable Parts Pose Estimation with Symmetry Annotation-Free Learning Strategy

TL;DR

This paper introduces SAFAG, a novel symmetry annotation-free framework for generalizable parts pose estimation that improves accuracy and robustness in data-lacking scenarios, advancing robot object interaction capabilities.

Contribution

SAFAG is the first framework to perform symmetry prediction without annotations, using a two-stage quaternion regression and self-supervised learning for improved parts pose estimation.

Findings

SAFAG outperforms existing methods in pose estimation accuracy.

SAFAG demonstrates robustness in data-lacking scenarios.

SAFAG's symmetry prediction enhances generalization in embodied AI.

Abstract

Urgently needed generalizable robot object interaction and manipulation requires high-quality Cross-Category object perception. As a pioneer of this area, Generalizable and Actionable Parts (GAParts) understanding has attracted increasing attention from relevant researchers. However, most recent works either have insufficient design regarding the symmetry issue or require rich symmetry annotation, which severely impedes precise GAPart pose estimation in data-lacking scenarios. In this paper, we propose SAFAG, a novel Symmetry Annotation-Free framework for Generalizable and Actionable Parts Pose Estimation. Specifically, we suggest a stepwise refinement two-stage framework for candidate-to-final quaternion regression, and tackle the symmetry prediction as a probability distribution problem with self-supervised learning strategy. The experimental results demonstrate the superior performance and robustness of our SAFAG. We believe that our work has the enormous potential to be applied in many areas of embodied AI system.