IMAGE TO POINT CLOUD TRANSLATION USING CONDITIONAL GENERATIVE ADVERSARIAL NETWORK FOR AIRBORNE LIDAR DATA
This study introduces a novel image to a 3D point-cloud translation method with a conditional generative adversarial network that creates a large-scale 3D point cloud. This can generate supervised point clouds observed via airborne LiDAR from aerial images. The network is composed of an encoder to p...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2021-06-01
|
| Series: | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/V-2-2021/169/2021/isprs-annals-V-2-2021-169-2021.pdf |
| _version_ | 1819064426122182656 |
|---|---|
| author | T. Shinohara H. Xiu M. Matsuoka |
| author_facet | T. Shinohara H. Xiu M. Matsuoka |
| author_sort | T. Shinohara |
| collection | DOAJ |
| description | This study introduces a novel image to a 3D point-cloud translation method with a conditional generative adversarial network that creates a large-scale 3D point cloud. This can generate supervised point clouds observed via airborne LiDAR from aerial images. The network is composed of an encoder to produce latent features of input images, generator to translate latent features to fake point clouds, and discriminator to classify false or real point clouds. The encoder is a pre-trained ResNet; to overcome the difficulty of generating 3D point clouds in an outdoor scene, we use a FoldingNet with features from ResNet. After a fixed number of iterations, our generator can produce fake point clouds that correspond to the input image. Experimental results show that our network can learn and generate certain point clouds using the data from the 2018 IEEE GRSS Data Fusion Contest. |
| first_indexed | 2024-12-21T15:30:23Z |
| format | Article |
| id | doaj.art-a3304b38f2a74df48c46a1052e275eca |
| institution | Directory Open Access Journal |
| issn | 2194-9042 2194-9050 |
| language | English |
| last_indexed | 2024-12-21T15:30:23Z |
| publishDate | 2021-06-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj.art-a3304b38f2a74df48c46a1052e275eca2022-12-21T18:58:46ZengCopernicus PublicationsISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences2194-90422194-90502021-06-01V-2-202116917410.5194/isprs-annals-V-2-2021-169-2021IMAGE TO POINT CLOUD TRANSLATION USING CONDITIONAL GENERATIVE ADVERSARIAL NETWORK FOR AIRBORNE LIDAR DATAT. Shinohara0H. Xiu1M. Matsuoka2Tokyo Institute of Technology, Department of Architecture and Building Engineering, Yokohama, JapanTokyo Institute of Technology, Department of Architecture and Building Engineering, Yokohama, JapanTokyo Institute of Technology, Department of Architecture and Building Engineering, Yokohama, JapanThis study introduces a novel image to a 3D point-cloud translation method with a conditional generative adversarial network that creates a large-scale 3D point cloud. This can generate supervised point clouds observed via airborne LiDAR from aerial images. The network is composed of an encoder to produce latent features of input images, generator to translate latent features to fake point clouds, and discriminator to classify false or real point clouds. The encoder is a pre-trained ResNet; to overcome the difficulty of generating 3D point clouds in an outdoor scene, we use a FoldingNet with features from ResNet. After a fixed number of iterations, our generator can produce fake point clouds that correspond to the input image. Experimental results show that our network can learn and generate certain point clouds using the data from the 2018 IEEE GRSS Data Fusion Contest.https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/V-2-2021/169/2021/isprs-annals-V-2-2021-169-2021.pdf |
| spellingShingle | T. Shinohara H. Xiu M. Matsuoka IMAGE TO POINT CLOUD TRANSLATION USING CONDITIONAL GENERATIVE ADVERSARIAL NETWORK FOR AIRBORNE LIDAR DATA ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | IMAGE TO POINT CLOUD TRANSLATION USING CONDITIONAL GENERATIVE ADVERSARIAL NETWORK FOR AIRBORNE LIDAR DATA |
| title_full | IMAGE TO POINT CLOUD TRANSLATION USING CONDITIONAL GENERATIVE ADVERSARIAL NETWORK FOR AIRBORNE LIDAR DATA |
| title_fullStr | IMAGE TO POINT CLOUD TRANSLATION USING CONDITIONAL GENERATIVE ADVERSARIAL NETWORK FOR AIRBORNE LIDAR DATA |
| title_full_unstemmed | IMAGE TO POINT CLOUD TRANSLATION USING CONDITIONAL GENERATIVE ADVERSARIAL NETWORK FOR AIRBORNE LIDAR DATA |
| title_short | IMAGE TO POINT CLOUD TRANSLATION USING CONDITIONAL GENERATIVE ADVERSARIAL NETWORK FOR AIRBORNE LIDAR DATA |
| title_sort | image to point cloud translation using conditional generative adversarial network for airborne lidar data |
| url | https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/V-2-2021/169/2021/isprs-annals-V-2-2021-169-2021.pdf |
| work_keys_str_mv | AT tshinohara imagetopointcloudtranslationusingconditionalgenerativeadversarialnetworkforairbornelidardata AT hxiu imagetopointcloudtranslationusingconditionalgenerativeadversarialnetworkforairbornelidardata AT mmatsuoka imagetopointcloudtranslationusingconditionalgenerativeadversarialnetworkforairbornelidardata |