3D Object Detection Based on Graph Network Fusion Sampling Strategy
In the 3D target detection technology based on point cloud, there are problems like high cost of point cloud calculation and large gap between target scales, which lead to low target detection efficiency. In response, this paper proposes a 3D object detection algorithm based on fusion sampling and g...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | Chinese |
| Published: |
Harbin University of Science and Technology Publications
2025-04-01
|
| Series: | Journal of Harbin University of Science and Technology |
| Subjects: | |
| Online Access: | https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2412 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839639976214528000 |
|---|---|
| author | LI Wenju CHEN Zhilin QU Jiantao CUI Liu CHU Wanghui GAO Hui |
| author_facet | LI Wenju CHEN Zhilin QU Jiantao CUI Liu CHU Wanghui GAO Hui |
| author_sort | LI Wenju |
| collection | DOAJ |
| description | In the 3D target detection technology based on point cloud, there are problems like high cost of point cloud calculation and large gap between target scales, which lead to low target detection efficiency. In response, this paper proposes a 3D object detection algorithm based on fusion sampling and graph networks. Firstly, the point cloud fusion sampling technology is introduced to sample the original point cloud to reduce the computational complexity. Secondly, the K-NN algorithm is used to construct the graph of the sampled point cloud, and sub-image sampling is introduced to solve the problem of over-smooth graph convolution. Finally, the features of graph nodes are updated through feature interaction to improve the feature extraction ability of the network, thereby improving the target detection effect. This study conducted experiments on the KITTI3D dataset. Compared with the benchmark model Point-GNN, the detection accuracy of the car target was improved by 3. 89%. In the simple scene and the medium difficulty scene, the detection accuracy of the cyclist target is increased by 6. 60% and 4. 36% , respectively. |
| format | Article |
| id | doaj-art-74040fabad22406bb79863c5a166da97 |
| institution | Matheson Library |
| issn | 1007-2683 |
| language | zho |
| publishDate | 2025-04-01 |
| publisher | Harbin University of Science and Technology Publications |
| record_format | Article |
| series | Journal of Harbin University of Science and Technology |
| spelling | doaj-art-74040fabad22406bb79863c5a166da972025-07-04T02:09:46ZzhoHarbin University of Science and Technology PublicationsJournal of Harbin University of Science and Technology1007-26832025-04-013002425210.15938/j.jhust.2025.02.0053D Object Detection Based on Graph Network Fusion Sampling StrategyLI Wenju0CHEN Zhilin1QU Jiantao2CUI Liu3CHU Wanghui4GAO Hui5School of Computer Science and Information Engineering, Shanghai Institute of Technology, Shanghai 201418, ChinaSchool of Computer Science and Information Engineering, Shanghai Institute of Technology, Shanghai 201418, ChinaShanghai Seagull Digital Camera Co. ,Ltd. , Shanghai 201613, ChinaSchool of Computer Science and Information Engineering, Shanghai Institute of Technology, Shanghai 201418, ChinaSchool of Computer Science and Information Engineering, Shanghai Institute of Technology, Shanghai 201418, ChinaSchool of Art and Design, Shanghai Institute of Technology, Shanghai 201418,ChinaIn the 3D target detection technology based on point cloud, there are problems like high cost of point cloud calculation and large gap between target scales, which lead to low target detection efficiency. In response, this paper proposes a 3D object detection algorithm based on fusion sampling and graph networks. Firstly, the point cloud fusion sampling technology is introduced to sample the original point cloud to reduce the computational complexity. Secondly, the K-NN algorithm is used to construct the graph of the sampled point cloud, and sub-image sampling is introduced to solve the problem of over-smooth graph convolution. Finally, the features of graph nodes are updated through feature interaction to improve the feature extraction ability of the network, thereby improving the target detection effect. This study conducted experiments on the KITTI3D dataset. Compared with the benchmark model Point-GNN, the detection accuracy of the car target was improved by 3. 89%. In the simple scene and the medium difficulty scene, the detection accuracy of the cyclist target is increased by 6. 60% and 4. 36% , respectively.https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2412point cloud3d object detectiongraph neural networkfusion samplingfeature fusion |
| spellingShingle | LI Wenju CHEN Zhilin QU Jiantao CUI Liu CHU Wanghui GAO Hui 3D Object Detection Based on Graph Network Fusion Sampling Strategy Journal of Harbin University of Science and Technology point cloud 3d object detection graph neural network fusion sampling feature fusion |
| title | 3D Object Detection Based on Graph Network Fusion Sampling Strategy |
| title_full | 3D Object Detection Based on Graph Network Fusion Sampling Strategy |
| title_fullStr | 3D Object Detection Based on Graph Network Fusion Sampling Strategy |
| title_full_unstemmed | 3D Object Detection Based on Graph Network Fusion Sampling Strategy |
| title_short | 3D Object Detection Based on Graph Network Fusion Sampling Strategy |
| title_sort | 3d object detection based on graph network fusion sampling strategy |
| topic | point cloud 3d object detection graph neural network fusion sampling feature fusion |
| url | https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2412 |
| work_keys_str_mv | AT liwenju 3dobjectdetectionbasedongraphnetworkfusionsamplingstrategy AT chenzhilin 3dobjectdetectionbasedongraphnetworkfusionsamplingstrategy AT qujiantao 3dobjectdetectionbasedongraphnetworkfusionsamplingstrategy AT cuiliu 3dobjectdetectionbasedongraphnetworkfusionsamplingstrategy AT chuwanghui 3dobjectdetectionbasedongraphnetworkfusionsamplingstrategy AT gaohui 3dobjectdetectionbasedongraphnetworkfusionsamplingstrategy |