SD-Net: Symmetric-Aware Keypoint Prediction and Domain Adaptation for 6D Pose Estimation In Bin-picking Scenarios

TL;DR

SD-Net introduces a symmetric-aware keypoint prediction and self-training domain adaptation framework that significantly improves 6D pose estimation accuracy for symmetric objects in bin-picking scenarios, addressing key challenges like ambiguity and domain gap.

Contribution

The paper presents a novel 6D pose estimation network with symmetry-aware keypoint prediction and self-training domain adaptation, achieving state-of-the-art results.

Findings

Achieves 96% average precision on Sil'eane dataset.

Outperforms previous methods by 8% on Parametric datasets.

Effectively handles symmetry ambiguity and domain gap challenges.

Abstract

Despite the success in 6D pose estimation in bin-picking scenarios, existing methods still struggle to produce accurate prediction results for symmetry objects and real world scenarios. The primary bottlenecks include 1) the ambiguity keypoints caused by object symmetries; 2) the domain gap between real and synthetic data. To circumvent these problem, we propose a new 6D pose estimation network with symmetric-aware keypoint prediction and self-training domain adaptation (SD-Net). SD-Net builds on pointwise keypoint regression and deep hough voting to perform reliable detection keypoint under clutter and occlusion. Specifically, at the keypoint prediction stage, we designe a robust 3D keypoints selection strategy considering the symmetry class of objects and equivalent keypoints, which facilitate locating 3D keypoints even in highly occluded scenes. Additionally, we build an effective filtering algorithm on predicted keypoint to dynamically eliminate multiple ambiguity and outlier keypoint candidates. At the domain adaptation stage, we propose the self-training framework using a student-teacher training scheme. To carefully distinguish reliable predictions, we harnesses a tailored heuristics for 3D geometry pseudo labelling based on semi-chamfer distance. On public Sil'eane dataset, SD-Net achieves state-of-the-art results, obtaining an average precision of 96%. Testing learning and generalization abilities on public Parametric datasets, SD-Net is 8% higher than the state-of-the-art method. The code is available at https://github.com/dingthuang/SD-Net.