Deep learning based point cloud registration: an overview
Point cloud registration aims at finding a rigid transformation to align one point cloud to another one. It is a fundamental problem in computer vision and robotics, which has been widely used in various applications, such as 3D reconstruction, SLAM (simultaneous localization and mapping), and auton...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co., Ltd.
2020-06-01
|
| Series: | Virtual Reality & Intelligent Hardware |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2096579620300383 |
| _version_ | 1828415276331302912 |
|---|---|
| author | Zhiyuan Zhang Yuchao Dai Jiadai Sun |
| author_facet | Zhiyuan Zhang Yuchao Dai Jiadai Sun |
| author_sort | Zhiyuan Zhang |
| collection | DOAJ |
| description | Point cloud registration aims at finding a rigid transformation to align one point cloud to another one. It is a fundamental problem in computer vision and robotics, which has been widely used in various applications, such as 3D reconstruction, SLAM (simultaneous localization and mapping), and autonomous driving. Over the last decades, many researchers have devoted themselves to tackle this challenging problem. Recently, the success of deep learning in high-level vision tasks has been extended to different geometric vision tasks. Various kinds of deep learning based point cloud registration methods have been proposed to exploit different aspects of the problem. However, a comprehensive overview of these approaches is still missing. To this end, in this paper, we summarize recent progress and present a comprehensive overview for deep learning based point cloud registration. We classify the popular approaches into different categories such as, correspondences-based or correspondences-free, effective modules: feature extractor, matching, outlier rejection, and motion estimation. Furthermore, we discuss the merits and demerits in detail. We provide a systematic and compact framework towards currently proposed methods and discuss future research directions. |
| first_indexed | 2024-12-10T13:41:56Z |
| format | Article |
| id | doaj.art-07b135ee65184f18bb881f77c9915314 |
| institution | Directory Open Access Journal |
| issn | 2096-5796 |
| language | English |
| last_indexed | 2024-12-10T13:41:56Z |
| publishDate | 2020-06-01 |
| publisher | KeAi Communications Co., Ltd. |
| record_format | Article |
| series | Virtual Reality & Intelligent Hardware |
| spelling | doaj.art-07b135ee65184f18bb881f77c99153142022-12-22T01:46:39ZengKeAi Communications Co., Ltd.Virtual Reality & Intelligent Hardware2096-57962020-06-0123222246Deep learning based point cloud registration: an overviewZhiyuan Zhang0Yuchao Dai1Jiadai Sun2School of Electronics and Information, Northwestern Polytechnical University, Shaanixi, 710129, ChinaSchool of Electronics and Information, Northwestern Polytechnical University, Shaanixi, 710129, China; Corresponding author.School of Computer Science and Technology, Northwestern Polytechnical University, Shaanixi, 710129, ChinaPoint cloud registration aims at finding a rigid transformation to align one point cloud to another one. It is a fundamental problem in computer vision and robotics, which has been widely used in various applications, such as 3D reconstruction, SLAM (simultaneous localization and mapping), and autonomous driving. Over the last decades, many researchers have devoted themselves to tackle this challenging problem. Recently, the success of deep learning in high-level vision tasks has been extended to different geometric vision tasks. Various kinds of deep learning based point cloud registration methods have been proposed to exploit different aspects of the problem. However, a comprehensive overview of these approaches is still missing. To this end, in this paper, we summarize recent progress and present a comprehensive overview for deep learning based point cloud registration. We classify the popular approaches into different categories such as, correspondences-based or correspondences-free, effective modules: feature extractor, matching, outlier rejection, and motion estimation. Furthermore, we discuss the merits and demerits in detail. We provide a systematic and compact framework towards currently proposed methods and discuss future research directions.http://www.sciencedirect.com/science/article/pii/S2096579620300383OverviewPoint Cloud RegistrationDeep LearningGraph neural networks |
| spellingShingle | Zhiyuan Zhang Yuchao Dai Jiadai Sun Deep learning based point cloud registration: an overview Virtual Reality & Intelligent Hardware Overview Point Cloud Registration Deep Learning Graph neural networks |
| title | Deep learning based point cloud registration: an overview |
| title_full | Deep learning based point cloud registration: an overview |
| title_fullStr | Deep learning based point cloud registration: an overview |
| title_full_unstemmed | Deep learning based point cloud registration: an overview |
| title_short | Deep learning based point cloud registration: an overview |
| title_sort | deep learning based point cloud registration an overview |
| topic | Overview Point Cloud Registration Deep Learning Graph neural networks |
| url | http://www.sciencedirect.com/science/article/pii/S2096579620300383 |
| work_keys_str_mv | AT zhiyuanzhang deeplearningbasedpointcloudregistrationanoverview AT yuchaodai deeplearningbasedpointcloudregistrationanoverview AT jiadaisun deeplearningbasedpointcloudregistrationanoverview |