Abstract

It is important to improve the registration precision and speed in the process of registration. In order to solve this problem, we proposed a robust point cloud registration method based on deep learning, called PDC-Net, using a principal component analysis based adjustment network that quickly adjusts the initial position between two slices of the point cloud, then using an iterative neural network based on the inverse compositional algorithm to complete the final registration transformation. We compare it on the ModelNet40 dataset with iterative closest point, which is the traditional point cloud registration method, and the learning-based methods including PointNet-LK and deep closest point. The experimental results show that the registration error is not worse with the increase of the initial phase between point clouds, avoiding the algorithm falling into the local optimal solution and enhancing the robustness of registration.

© 2021 Optical Society of America

ICP registration with DCA descriptor for 3D point clouds

Ying He, Jun Yang, Xingming Hou, Shiyan Pang, and Jia Chen
Opt. Express 29(13) 20423-20439 (2021)

Registration of 3D point clouds using a local descriptor based on grid point normal

Jiang Wang, Bin Wu, and Jiehu Kang
Appl. Opt. 60(28) 8818-8828 (2021)

Hierarchical registration of laser point clouds between airborne and vehicle-borne data considering building eave attributes

Yadong Guo, Xiankun Wang, Dianpeng Su, Fanlin Yang, Guoyu Li, and Chao Qi
Appl. Opt. 60(15) C20-C31 (2021)

Multi-level height maps-based registration method for sparse LiDAR point clouds in an urban scene

Bin Fang, Jie Ma, Pei An, Zhao Wang, Jun Zhang, and Kun Yu
Appl. Opt. 60(14) 4154-4164 (2021)

Simple and efficient registration of 3D point cloud and image data for an indoor mobile mapping system

Hao Ma, Keke Liu, Jingbin Liu, Hongyu Qiu, Dong Xu, Zemin Wang, Xiaodong Gong, and Sheng Yang
J. Opt. Soc. Am. A 38(4) 579-586 (2021)