GTPCR: Graph-Enhanced Transformer for Point Cloud Registration

• Published in: 2024 27th International Conference on Computer Supported Cooperative Work in Design (CSCWD) pp. 1304 - 1309
• Main Authors: Chen, Kai, Yao, Junfeng, Li, Yuanhang, Zhang, Han, Shen, Huabo, Qian, Quan, Wu, Xing
• Format: Conference Proceeding
• Language: English
• Published: IEEE 08.05.2024
• Subjects: Collaborative Detection; Defect Detection; Graph Enhancement; Network architecture; Point cloud compression; Point Cloud Registration; Scalability; Shape; Solid modeling; Three-dimensional displays; Transformers
• ISSN: 2768-1904
• DOI: 10.1109/CSCWD61410.2024.10580723

As Industry 4.0 continues to advance, point cloud registration technology is extensively employed in scenarios such as collaborative defect detection in products and digital twin-assisted assembly. In this paper, we present an end-to-end point cloud registration model, GTPCR, which is based on the utilization of spatial structural information within point cloud data. GTPCR employs a point-wise approach, directly solving rigid rotations based on estimated correspondences without reliance on RANSAC. It conceives of the point cloud as a graph structure in three-dimensional space, encoding nodes across various dimensions. The encoding of an individual node is dictated by its position and centrality. The geometric associations between nodes are contingent on factors such as relative position, Euclidean distance, azimuth, and elevation. Edge feature encoding is dynamically acquired through the learning process from node features. All encodings are incorporated as trainable biases input to the model. In comparison to methods transitioning from local to global, GTPCR demonstrates superior efficiency and heightened scalability. Moreover, due to its adept network architecture design, GTPCR effortlessly expands its applicability to non-rigid registration. Empirical evidence undeniably illustrates GTPCR's competitive advantage across numerous datasets. In particular, GTPCR shows significant improvements in registration recall of 1.4% and 0.9% over baselines on the 3DMatch and 3DLoMatch datasets, respectively.