Hybrid optimization with unconstrained variables on partial point cloud registration. (April 2023)

• Record Type: Journal Article
• Authors: Yan, Yuanjie; An, Junyi; Zhao, Jian; Shen, Furao
• Is Part Of: Pattern recognition. Volume 136(2023)
• Publication Date: 2023-04
• Subjects: Point cloud registration -- Optimization; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2022.109267
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We propose the novel hybrid optimization method with unconstrained variables on pairwise partial point cloud registration. To address the local and global point matching, we propose the local CD loss and projected CD loss based on chamfer distance. The proposed optimization strategies are programmable and flexible to fit the sample or dataset. Our method won the first place in the multi-view partial point clouds registration in ICCV 2021. We also verify the performance of the proposed methods on the synthetic and realistic datasets. Abstract: 3D point cloud registration is a fundamental problem in computer vision (CV) and computer graphics (CG). Recently, a series of learning-based algorithms have been proposed to show the advantages in registration accuracy and inference speed. However, those learning-based methods usually ignore transformations with constrained rotations and translations in registration. In this paper, we propose a novel hybrid optimization method to solve the constrained rotational and translational transformations. A mapping function is introduced to deal with the restrained variables in optimization. Our method achieves superior performance on the Multi-View Partial Point dataset, which won the first place on the registration challenge in ICCV 2021. The method is also validated on the synthetic datasets ModelNet, ICL-NUIM, and the realistic 3DMatch dataset. We demonstrate that the global optimization methods still have great potential research for point cloud registration. The code is available at https://github.com/Dizzy-cell/HOUV .