GACM: A Graph Attention Capsule Model for the Registration of TLS Point Clouds in the Urban Scene
Point cloud registration is the foundation and key step for many vital applications, such as digital city, autonomous driving, passive positioning, and navigation. The difference of spatial objects and the structure complexity of object surfaces are the main challenges for the registration problem....
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-11-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/22/4497 |
| _version_ | 1827675499118198784 |
|---|---|
| author | Jianjun Zou Zhenxin Zhang Dong Chen Qinghua Li Lan Sun Ruofei Zhong Liqiang Zhang Jinghan Sha |
| author_facet | Jianjun Zou Zhenxin Zhang Dong Chen Qinghua Li Lan Sun Ruofei Zhong Liqiang Zhang Jinghan Sha |
| author_sort | Jianjun Zou |
| collection | DOAJ |
| description | Point cloud registration is the foundation and key step for many vital applications, such as digital city, autonomous driving, passive positioning, and navigation. The difference of spatial objects and the structure complexity of object surfaces are the main challenges for the registration problem. In this paper, we propose a graph attention capsule model (named as GACM) for the efficient registration of terrestrial laser scanning (TLS) point cloud in the urban scene, which fuses graph attention convolution and a three-dimensional (3D) capsule network to extract local point cloud features and obtain 3D feature descriptors. These descriptors can take into account the differences of spatial structure and point density in objects and make the spatial features of ground objects more prominent. During the training progress, we used both matched points and non-matched points to train the model. In the test process of the registration, the points in the neighborhood of each keypoint were sent to the trained network, in order to obtain feature descriptors and calculate the rotation and translation matrix after constructing a K-dimensional (KD) tree and random sample consensus (RANSAC) algorithm. Experiments show that the proposed method achieves more efficient registration results and higher robustness than other frontier registration methods in the pairwise registration of point clouds. |
| first_indexed | 2024-03-10T05:05:50Z |
| format | Article |
| id | doaj.art-3bceed96fb7c41729228d5c11d7d95af |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T05:05:50Z |
| publishDate | 2021-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-3bceed96fb7c41729228d5c11d7d95af2023-11-23T01:18:15ZengMDPI AGRemote Sensing2072-42922021-11-011322449710.3390/rs13224497GACM: A Graph Attention Capsule Model for the Registration of TLS Point Clouds in the Urban SceneJianjun Zou0Zhenxin Zhang1Dong Chen2Qinghua Li3Lan Sun4Ruofei Zhong5Liqiang Zhang6Jinghan Sha7Key Laboratory of 3D Information Acquisition and Application, MOE, Capital Normal University, Beijing 100048, ChinaKey Laboratory of 3D Information Acquisition and Application, MOE, Capital Normal University, Beijing 100048, ChinaCollege of Civil Engineering, Nanjing Forestry University, Nanjing 210037, ChinaNorthwestern Institute on Complex Systems, Northwestern University, Evanston, IL 60201, USAKey Laboratory of 3D Information Acquisition and Application, MOE, Capital Normal University, Beijing 100048, ChinaKey Laboratory of 3D Information Acquisition and Application, MOE, Capital Normal University, Beijing 100048, ChinaState Key Laboratory of Remote Sensing Science, Beijing Normal University, Beijing 100875, ChinaKey Laboratory of 3D Information Acquisition and Application, MOE, Capital Normal University, Beijing 100048, ChinaPoint cloud registration is the foundation and key step for many vital applications, such as digital city, autonomous driving, passive positioning, and navigation. The difference of spatial objects and the structure complexity of object surfaces are the main challenges for the registration problem. In this paper, we propose a graph attention capsule model (named as GACM) for the efficient registration of terrestrial laser scanning (TLS) point cloud in the urban scene, which fuses graph attention convolution and a three-dimensional (3D) capsule network to extract local point cloud features and obtain 3D feature descriptors. These descriptors can take into account the differences of spatial structure and point density in objects and make the spatial features of ground objects more prominent. During the training progress, we used both matched points and non-matched points to train the model. In the test process of the registration, the points in the neighborhood of each keypoint were sent to the trained network, in order to obtain feature descriptors and calculate the rotation and translation matrix after constructing a K-dimensional (KD) tree and random sample consensus (RANSAC) algorithm. Experiments show that the proposed method achieves more efficient registration results and higher robustness than other frontier registration methods in the pairwise registration of point clouds.https://www.mdpi.com/2072-4292/13/22/4497TLS point cloud registrationurban sceneGACMgraph attentioncapsule network |
| spellingShingle | Jianjun Zou Zhenxin Zhang Dong Chen Qinghua Li Lan Sun Ruofei Zhong Liqiang Zhang Jinghan Sha GACM: A Graph Attention Capsule Model for the Registration of TLS Point Clouds in the Urban Scene Remote Sensing TLS point cloud registration urban scene GACM graph attention capsule network |
| title | GACM: A Graph Attention Capsule Model for the Registration of TLS Point Clouds in the Urban Scene |
| title_full | GACM: A Graph Attention Capsule Model for the Registration of TLS Point Clouds in the Urban Scene |
| title_fullStr | GACM: A Graph Attention Capsule Model for the Registration of TLS Point Clouds in the Urban Scene |
| title_full_unstemmed | GACM: A Graph Attention Capsule Model for the Registration of TLS Point Clouds in the Urban Scene |
| title_short | GACM: A Graph Attention Capsule Model for the Registration of TLS Point Clouds in the Urban Scene |
| title_sort | gacm a graph attention capsule model for the registration of tls point clouds in the urban scene |
| topic | TLS point cloud registration urban scene GACM graph attention capsule network |
| url | https://www.mdpi.com/2072-4292/13/22/4497 |
| work_keys_str_mv | AT jianjunzou gacmagraphattentioncapsulemodelfortheregistrationoftlspointcloudsintheurbanscene AT zhenxinzhang gacmagraphattentioncapsulemodelfortheregistrationoftlspointcloudsintheurbanscene AT dongchen gacmagraphattentioncapsulemodelfortheregistrationoftlspointcloudsintheurbanscene AT qinghuali gacmagraphattentioncapsulemodelfortheregistrationoftlspointcloudsintheurbanscene AT lansun gacmagraphattentioncapsulemodelfortheregistrationoftlspointcloudsintheurbanscene AT ruofeizhong gacmagraphattentioncapsulemodelfortheregistrationoftlspointcloudsintheurbanscene AT liqiangzhang gacmagraphattentioncapsulemodelfortheregistrationoftlspointcloudsintheurbanscene AT jinghansha gacmagraphattentioncapsulemodelfortheregistrationoftlspointcloudsintheurbanscene |