A review of rigid point cloud registration based on deep learning
With the development of 3D scanning devices, point cloud registration is gradually being applied in various fields. Traditional point cloud registration methods face challenges in noise, low overlap, uneven density, and large data scale, which limits the further application of point cloud registrati...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-01-01
|
| Series: | Frontiers in Neurorobotics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnbot.2023.1281332/full |
| _version_ | 1797366754986426368 |
|---|---|
| author | Lei Chen Changzhou Feng Yunpeng Ma Yikai Zhao Chaorong Wang |
| author_facet | Lei Chen Changzhou Feng Yunpeng Ma Yikai Zhao Chaorong Wang |
| author_sort | Lei Chen |
| collection | DOAJ |
| description | With the development of 3D scanning devices, point cloud registration is gradually being applied in various fields. Traditional point cloud registration methods face challenges in noise, low overlap, uneven density, and large data scale, which limits the further application of point cloud registration in actual scenes. With the above deficiency, point cloud registration methods based on deep learning technology gradually emerged. This review summarizes the point cloud registration technology based on deep learning. Firstly, point cloud registration based on deep learning can be categorized into two types: complete overlap point cloud registration and partially overlapping point cloud registration. And the characteristics of the two kinds of methods are classified and summarized in detail. The characteristics of the partially overlapping point cloud registration method are introduced and compared with the completely overlapping method to provide further research insight. Secondly, the review delves into network performance improvement summarizes how to accelerate the point cloud registration method of deep learning from the hardware and software. Then, this review discusses point cloud registration applications in various domains. Finally, this review summarizes and outlooks the current challenges and future research directions of deep learning-based point cloud registration. |
| first_indexed | 2024-03-08T17:08:00Z |
| format | Article |
| id | doaj.art-ce2f7ce3d40c4560bcfc3c27e80df073 |
| institution | Directory Open Access Journal |
| issn | 1662-5218 |
| language | English |
| last_indexed | 2024-03-08T17:08:00Z |
| publishDate | 2024-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurorobotics |
| spelling | doaj.art-ce2f7ce3d40c4560bcfc3c27e80df0732024-01-04T04:33:27ZengFrontiers Media S.A.Frontiers in Neurorobotics1662-52182024-01-011710.3389/fnbot.2023.12813321281332A review of rigid point cloud registration based on deep learningLei ChenChangzhou FengYunpeng MaYikai ZhaoChaorong WangWith the development of 3D scanning devices, point cloud registration is gradually being applied in various fields. Traditional point cloud registration methods face challenges in noise, low overlap, uneven density, and large data scale, which limits the further application of point cloud registration in actual scenes. With the above deficiency, point cloud registration methods based on deep learning technology gradually emerged. This review summarizes the point cloud registration technology based on deep learning. Firstly, point cloud registration based on deep learning can be categorized into two types: complete overlap point cloud registration and partially overlapping point cloud registration. And the characteristics of the two kinds of methods are classified and summarized in detail. The characteristics of the partially overlapping point cloud registration method are introduced and compared with the completely overlapping method to provide further research insight. Secondly, the review delves into network performance improvement summarizes how to accelerate the point cloud registration method of deep learning from the hardware and software. Then, this review discusses point cloud registration applications in various domains. Finally, this review summarizes and outlooks the current challenges and future research directions of deep learning-based point cloud registration.https://www.frontiersin.org/articles/10.3389/fnbot.2023.1281332/fullpoint cloud registrationdeep learningpartial overlapnetwork accelerationneural networks |
| spellingShingle | Lei Chen Changzhou Feng Yunpeng Ma Yikai Zhao Chaorong Wang A review of rigid point cloud registration based on deep learning Frontiers in Neurorobotics point cloud registration deep learning partial overlap network acceleration neural networks |
| title | A review of rigid point cloud registration based on deep learning |
| title_full | A review of rigid point cloud registration based on deep learning |
| title_fullStr | A review of rigid point cloud registration based on deep learning |
| title_full_unstemmed | A review of rigid point cloud registration based on deep learning |
| title_short | A review of rigid point cloud registration based on deep learning |
| title_sort | review of rigid point cloud registration based on deep learning |
| topic | point cloud registration deep learning partial overlap network acceleration neural networks |
| url | https://www.frontiersin.org/articles/10.3389/fnbot.2023.1281332/full |
| work_keys_str_mv | AT leichen areviewofrigidpointcloudregistrationbasedondeeplearning AT changzhoufeng areviewofrigidpointcloudregistrationbasedondeeplearning AT yunpengma areviewofrigidpointcloudregistrationbasedondeeplearning AT yikaizhao areviewofrigidpointcloudregistrationbasedondeeplearning AT chaorongwang areviewofrigidpointcloudregistrationbasedondeeplearning AT leichen reviewofrigidpointcloudregistrationbasedondeeplearning AT changzhoufeng reviewofrigidpointcloudregistrationbasedondeeplearning AT yunpengma reviewofrigidpointcloudregistrationbasedondeeplearning AT yikaizhao reviewofrigidpointcloudregistrationbasedondeeplearning AT chaorongwang reviewofrigidpointcloudregistrationbasedondeeplearning |