3DSAC: Size Adaptive Clustering for 3D object detection in point clouds
3D object detection is important for various indoor applications to understand the environment. Previous voting-based methods voted on the center of each seed point, which may suffer from errors from background points or adjacent objects. And the size-fixed feature grouping module is unsuitable for...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-04-01
|
| Series: | International Journal of Applied Earth Observations and Geoinformation |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1569843223000535 |
| _version_ | 1797842814872059904 |
|---|---|
| author | Hang Yu Jinhe Su Guorong Cai Yingchao Piao Niansheng Liu Min Huang |
| author_facet | Hang Yu Jinhe Su Guorong Cai Yingchao Piao Niansheng Liu Min Huang |
| author_sort | Hang Yu |
| collection | DOAJ |
| description | 3D object detection is important for various indoor applications to understand the environment. Previous voting-based methods voted on the center of each seed point, which may suffer from errors from background points or adjacent objects. And the size-fixed feature grouping module is unsuitable for indoor objects with variable sizes. In this paper, we propose a Size Adaptive Clustering method for 3D object detection in point clouds . First, we present a super-voting module to divide seed points into foreground and background points and perform enhanced voting on the foreground seeds. To create a good match for the feature clustering area and the size of an object, we design a size-adaptive clustering module to infer a clustering radius based on the seed-to-vote displacement offset. Finally, because indoor objects are highly related to spatial room layouts, a position-aware module is used to calculate aware weights among objects and enhance the features of occluded objects. Experiments show that our method outperforms VoteNet by a large margin on ScanNet V2 (mAP@0.250 +8.3%, mAP@0.50 +14.2%) and SUN RGB-D (mAP@0.250 +3.5%, mAP@0.50 +13.6%). The proposed method can detect indoor objects with variable sizes in high accuracy, and perform robustly in case of occluded objects. The code of 3DSAC will be available at github-3DSAC. |
| first_indexed | 2024-04-09T16:54:22Z |
| format | Article |
| id | doaj.art-051e3d26e2244afe9fee1ce6ce9a7fd9 |
| institution | Directory Open Access Journal |
| issn | 1569-8432 |
| language | English |
| last_indexed | 2024-04-09T16:54:22Z |
| publishDate | 2023-04-01 |
| publisher | Elsevier |
| record_format | Article |
| series | International Journal of Applied Earth Observations and Geoinformation |
| spelling | doaj.art-051e3d26e2244afe9fee1ce6ce9a7fd92023-04-21T06:41:05ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322023-04-011181032313DSAC: Size Adaptive Clustering for 3D object detection in point cloudsHang Yu0Jinhe Su1Guorong Cai2Yingchao Piao3Niansheng Liu4Min Huang5The School of Computer Engineering, Jimei University, Xiamen 361021, ChinaThe School of Computer Engineering, Jimei University, Xiamen 361021, China; Corresponding authors.The School of Computer Engineering, Jimei University, Xiamen 361021, ChinaThe Computer Network Information Center, Chinese Academy of Sciences, Beijing, ChinaThe School of Computer Engineering, Jimei University, Xiamen 361021, ChinaThe School of Computer Engineering, Jimei University, Xiamen 361021, China; Corresponding authors.3D object detection is important for various indoor applications to understand the environment. Previous voting-based methods voted on the center of each seed point, which may suffer from errors from background points or adjacent objects. And the size-fixed feature grouping module is unsuitable for indoor objects with variable sizes. In this paper, we propose a Size Adaptive Clustering method for 3D object detection in point clouds . First, we present a super-voting module to divide seed points into foreground and background points and perform enhanced voting on the foreground seeds. To create a good match for the feature clustering area and the size of an object, we design a size-adaptive clustering module to infer a clustering radius based on the seed-to-vote displacement offset. Finally, because indoor objects are highly related to spatial room layouts, a position-aware module is used to calculate aware weights among objects and enhance the features of occluded objects. Experiments show that our method outperforms VoteNet by a large margin on ScanNet V2 (mAP@0.250 +8.3%, mAP@0.50 +14.2%) and SUN RGB-D (mAP@0.250 +3.5%, mAP@0.50 +13.6%). The proposed method can detect indoor objects with variable sizes in high accuracy, and perform robustly in case of occluded objects. The code of 3DSAC will be available at github-3DSAC.http://www.sciencedirect.com/science/article/pii/S15698432230005353D object detectionPoint cloudHough votingSize adaptiveRelation information |
| spellingShingle | Hang Yu Jinhe Su Guorong Cai Yingchao Piao Niansheng Liu Min Huang 3DSAC: Size Adaptive Clustering for 3D object detection in point clouds International Journal of Applied Earth Observations and Geoinformation 3D object detection Point cloud Hough voting Size adaptive Relation information |
| title | 3DSAC: Size Adaptive Clustering for 3D object detection in point clouds |
| title_full | 3DSAC: Size Adaptive Clustering for 3D object detection in point clouds |
| title_fullStr | 3DSAC: Size Adaptive Clustering for 3D object detection in point clouds |
| title_full_unstemmed | 3DSAC: Size Adaptive Clustering for 3D object detection in point clouds |
| title_short | 3DSAC: Size Adaptive Clustering for 3D object detection in point clouds |
| title_sort | 3dsac size adaptive clustering for 3d object detection in point clouds |
| topic | 3D object detection Point cloud Hough voting Size adaptive Relation information |
| url | http://www.sciencedirect.com/science/article/pii/S1569843223000535 |
| work_keys_str_mv | AT hangyu 3dsacsizeadaptiveclusteringfor3dobjectdetectioninpointclouds AT jinhesu 3dsacsizeadaptiveclusteringfor3dobjectdetectioninpointclouds AT guorongcai 3dsacsizeadaptiveclusteringfor3dobjectdetectioninpointclouds AT yingchaopiao 3dsacsizeadaptiveclusteringfor3dobjectdetectioninpointclouds AT nianshengliu 3dsacsizeadaptiveclusteringfor3dobjectdetectioninpointclouds AT minhuang 3dsacsizeadaptiveclusteringfor3dobjectdetectioninpointclouds |