Sparse-to-Dense Matching Network for Large-Scale LiDAR Point Cloud Registration
IEEE Transactions on Pattern Analysis and Machine Intelligence · 01.09.2023 · 45 (9) · 11270-11282
DOI: 10.1109/tpami.2023.3265531 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, erschienen am 01.09.2023. Verfasst von Fan Lu und 6 weiteren Autor:innen. Die Arbeit wurde bisher 37-mal zitiert (Crossref) und nennt 62 Literaturangaben. Gefördert von: Shanghai Municipal Science and Technology Major Project, ZJ Lab, Shanghai Center for Brain Science and Brain-Inspired Technology.
Abstract (Originalsprache)
Point cloud registration is a fundamental problem in 3D computer vision. Previous learning-based methods for LiDAR point cloud registration can be categorized into two schemes: dense-to-dense matching methods and sparse-to-sparse matching methods. However, for large-scale outdoor LiDAR point clouds, solving dense point correspondences is time-consuming, whereas sparse keypoint matching easily suffers from keypoint detection error. In this paper, we propose SDMNet, a novel Sparse-to-Dense Matching Network for large-scale outdoor LiDAR point cloud registration. Specifically, SDMNet performs registration in two sequential stages: sparse matching stage and local-dense matching stage. In the sparse matching stage, we sample a set of sparse points from the source point cloud and then match them to the dense target point cloud using a spatial consistency enhanced soft matching network and a robust outlier rejection module. Furthermore, a novel neighborhood matching module is developed to incorporate local neighborhood consensus, significantly improving performance. The local-dense matching stage is followed for fine-grained performance, where dense correspondences are efficiently obtained by performing point matching in local spatial neighborhoods of high-confidence sparse correspondences. Extensive experiments on three large-scale outdoor LiDAR point cloud datasets demonstrate that the proposed SDMNet achieves state-of-the-art performance with high efficiency.
Förderung
- Shanghai Municipal Science and Technology Major Project
- ZJ Lab
- Shanghai Center for Brain Science and Brain-Inspired Technology
- Shanghai Rising Star Program
- Tongji-Westwell Autonomous Vehicle Joint Lab Project
- Key Technologies Development and Application of Piloted Autonomous Driving Truck Project
Zitieren
APA
Lu, F., Chen, G., Liu, Y., Zhan, Y., Li, Z., Tao, D., & Jiang, C. (2023). Sparse-to-Dense Matching Network for Large-Scale LiDAR Point Cloud Registration. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(9), 11270-11282. https://doi.org/10.1109/tpami.2023.3265531
Vancouver
Lu F, Chen G, Liu Y, Zhan Y, Li Z, Tao D, et al. Sparse-to-Dense Matching Network for Large-Scale LiDAR Point Cloud Registration. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2023;45(9):11270-11282. doi:10.1109/tpami.2023.3265531
BibTeX
@article{Lu2023,
author = {Lu, Fan and Chen, Guang and Liu, Yinlong and Zhan, Yibing and Li, Zhijun and Tao, Dacheng and Jiang, Changjun},
title = {Sparse-to-Dense Matching Network for Large-Scale LiDAR Point Cloud Registration},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2023},
volume = {45},
number = {9},
pages = {11270--11282},
doi = {10.1109/tpami.2023.3265531}
}
EUVIMED