3D point cloud generation reconstruction from single image based on image retrieval
Most of the 3D point cloud generation methods from a single image are only applicable to a single target. However, the real image often contains multiple targets, which is difficult to generate 3D point clouds from the network directly. To address this problem, this work proposes an end-to-end point...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-12-01
|
| Series: | Results in Optics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666950121000729 |
| _version_ | 1819319207235420160 |
|---|---|
| author | Hui Chen Yipeng Zuo Yong Tong Li Zhu |
| author_facet | Hui Chen Yipeng Zuo Yong Tong Li Zhu |
| author_sort | Hui Chen |
| collection | DOAJ |
| description | Most of the 3D point cloud generation methods from a single image are only applicable to a single target. However, the real image often contains multiple targets, which is difficult to generate 3D point clouds from the network directly. To address this problem, this work proposes an end-to-end point cloud generation network that integrates three parts: image cropping, image retrieval and point cloud reconstruction. The addition of the image cropping module allows network to deal with multi-target situations that appear in image. The image retrieval module can retrieve similar images in the ShapeNet dataset, and input their corresponding ground truth point cloud and target images to the reconstruction network to obtain the three-dimensional reconstruction model. Experiments on a single image containing single and multiple targets show that our method achieves point cloud reconstruction of real image and obtains a more accurate three-dimensional model. |
| first_indexed | 2024-12-24T11:00:01Z |
| format | Article |
| id | doaj.art-7ebe4b24a78a4bc0945b45df246fbd4e |
| institution | Directory Open Access Journal |
| issn | 2666-9501 |
| language | English |
| last_indexed | 2024-12-24T11:00:01Z |
| publishDate | 2021-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Results in Optics |
| spelling | doaj.art-7ebe4b24a78a4bc0945b45df246fbd4e2022-12-21T16:58:44ZengElsevierResults in Optics2666-95012021-12-0151001243D point cloud generation reconstruction from single image based on image retrievalHui Chen0Yipeng Zuo1Yong Tong2Li Zhu3Corresponding author.; College of Automation Engineering, Shanghai University of Electric Power, Shanghai 310027, ChinaCollege of Automation Engineering, Shanghai University of Electric Power, Shanghai 310027, ChinaCollege of Automation Engineering, Shanghai University of Electric Power, Shanghai 310027, ChinaCollege of Automation Engineering, Shanghai University of Electric Power, Shanghai 310027, ChinaMost of the 3D point cloud generation methods from a single image are only applicable to a single target. However, the real image often contains multiple targets, which is difficult to generate 3D point clouds from the network directly. To address this problem, this work proposes an end-to-end point cloud generation network that integrates three parts: image cropping, image retrieval and point cloud reconstruction. The addition of the image cropping module allows network to deal with multi-target situations that appear in image. The image retrieval module can retrieve similar images in the ShapeNet dataset, and input their corresponding ground truth point cloud and target images to the reconstruction network to obtain the three-dimensional reconstruction model. Experiments on a single image containing single and multiple targets show that our method achieves point cloud reconstruction of real image and obtains a more accurate three-dimensional model.http://www.sciencedirect.com/science/article/pii/S2666950121000729Deep learningPoint cloudMulti-target3d reconstruction |
| spellingShingle | Hui Chen Yipeng Zuo Yong Tong Li Zhu 3D point cloud generation reconstruction from single image based on image retrieval Results in Optics Deep learning Point cloud Multi-target 3d reconstruction |
| title | 3D point cloud generation reconstruction from single image based on image retrieval |
| title_full | 3D point cloud generation reconstruction from single image based on image retrieval |
| title_fullStr | 3D point cloud generation reconstruction from single image based on image retrieval |
| title_full_unstemmed | 3D point cloud generation reconstruction from single image based on image retrieval |
| title_short | 3D point cloud generation reconstruction from single image based on image retrieval |
| title_sort | 3d point cloud generation reconstruction from single image based on image retrieval |
| topic | Deep learning Point cloud Multi-target 3d reconstruction |
| url | http://www.sciencedirect.com/science/article/pii/S2666950121000729 |
| work_keys_str_mv | AT huichen 3dpointcloudgenerationreconstructionfromsingleimagebasedonimageretrieval AT yipengzuo 3dpointcloudgenerationreconstructionfromsingleimagebasedonimageretrieval AT yongtong 3dpointcloudgenerationreconstructionfromsingleimagebasedonimageretrieval AT lizhu 3dpointcloudgenerationreconstructionfromsingleimagebasedonimageretrieval |