Robust Point Cloud Registration Framework Based on Deep Graph Matching

• Published in: IEEE transactions on pattern analysis and machine intelligence 2023-05, Vol.45 (5), p.6183-6195
• Main Authors: Fu, Kexue, Luo, Jiazheng, Luo, Xiaoyuan, Liu, Shaolei, Zhang, Chenxi, Wang, Manning
• Format: Article
• Language: English
• Subjects: Computer vision; Correspondence; Deep learning; Feature extraction; Geometry; Graph matching; Graph theory; Neural networks; Outliers (statistics); Point cloud compression; point cloud registration; Prediction algorithms; Registration; Robotics; Three dimensional models; Topology; Transformers
• ISSN: 0162-8828; 1939-3539; 2160-9292
• DOI: 10.1109/TPAMI.2022.3204713

3D point cloud registration is a fundamental problem in computer vision and robotics. Recently, learning-based point cloud registration methods have made great progress. However, these methods are sensitive to outliers, which lead to more incorrect correspondences. In this paper, we propose a novel deep graph matching-based framework for point cloud registration. Specifically, we first transform point clouds into graphs and extract deep features for each point. Then, we develop a module based on deep graph matching to calculate a soft correspondence matrix. By using graph matching, not only the local geometry of each point but also its structure and topology in a larger range are considered in establishing correspondences, so that more correct correspondences are found. We train the network with a loss directly defined on the correspondences, and in the test stage the soft correspondences are transformed into hard one-to-one correspondences so that registration can be performed by a correspondence-based solver. Furthermore, we introduce a transformer-based method to generate edges for graph construction, which further improves the quality of the correspondences. Extensive experiments on object-level and scene-level benchmark datasets show that the proposed method achieves state-of-the-art performance.