Benchmarking Classical and Learning-Based Multibeam Point Cloud Registration

• Published in: 2024 IEEE International Conference on Robotics and Automation (ICRA) pp. 6118 - 6125
• Main Authors: Ling, Li, Zhang, Jun, Bore, Nils, Folkesson, John, Wahlin, Anna
• Format: Conference Proceeding
• Language: English
• Published: IEEE 13.05.2024
• Subjects: Computer and Information Sciences; Data- och informationsvetenskap (Datateknik); Iterative closest point algorithm; Learning systems; Measurement; Point cloud compression; Simultaneous localization and mapping; Three-dimensional displays; Transforms
• ISBN: 9798350384574
• ISSN: 1050-4729
• DOI: 10.1109/ICRA57147.2024.10610118

Deep learning has shown promising results for multiple 3D point cloud registration datasets. However, in the underwater domain, most registration of multibeam echo-sounder (MBES) point cloud data are still performed using classical methods in the iterative closest point (ICP) family. In this work, we curate and release DotsonEast Dataset, a semi-synthetic MBES registration dataset constructed from an autonomous underwater vehicle in West Antarctica. Using this dataset, we systematically benchmark the performance of 2 classical and 4 learning-based methods. The experimental results show that the learning-based methods work well for coarse alignment, and are better at recovering rough transforms consistently at high overlap (20-50%). In comparison, GICP (a variant of ICP) performs well for fine alignment and is better across all metrics at extremely low overlap (10%). To the best of our knowledge, this is the first work to benchmark both learning-based and classical registration methods on an AUV-based MBES dataset. To facilitate future research, both the code and data are made available online. 1