SPBA-Net point cloud object detection with sparse attention and box aligning
Abstract Object detection in point clouds is essential for various applications, including autonomous navigation, household robots, and augmented/virtual reality. However, during voxelization and Bird’s Eye View transformation, local point cloud data often remains sparse due to non-target areas and...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-11-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-77097-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846158701184942080 |
|---|---|
| author | Haojie Sha Qingrui Gao Hao Zeng Kai Li Wang Li Xuande Zhang Xiaohui Wang |
| author_facet | Haojie Sha Qingrui Gao Hao Zeng Kai Li Wang Li Xuande Zhang Xiaohui Wang |
| author_sort | Haojie Sha |
| collection | DOAJ |
| description | Abstract Object detection in point clouds is essential for various applications, including autonomous navigation, household robots, and augmented/virtual reality. However, during voxelization and Bird’s Eye View transformation, local point cloud data often remains sparse due to non-target areas and noise points, posing a significant challenge for feature extraction. In this paper, we propose a novel mechanism named Keypoint Guided Sparse Attention (KGSA) to enhance the semantic information of point clouds by calculating Euclidean distances between selected keypoints and others. Additionally, we introduce Instance-wise Box Aligning, a method for expanding predicted boxes and clustering the points within them to achieve precise alignment between predicted bounding boxes and ground-truth targets. Experimental results demonstrate the superiority of our proposed SPBA-Net in 3D object detection on point clouds compared to other state-of-the-art methods.The datasets used and/or analysed during the current study available from the corresponding author on reasonable request. |
| format | Article |
| id | doaj-art-dca67bd8dc324770b0ad2bb38d0b088f |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-dca67bd8dc324770b0ad2bb38d0b088f2024-11-24T12:18:24ZengNature PortfolioScientific Reports2045-23222024-11-0114111010.1038/s41598-024-77097-zSPBA-Net point cloud object detection with sparse attention and box aligningHaojie Sha0Qingrui Gao1Hao Zeng2Kai Li3Wang Li4Xuande Zhang5Xiaohui Wang6College of Applied Technology, Qingdao UniversityCollege of Applied Technology, Qingdao UniversityInstitute of Software Chinese Academy of SciencesCollege of Applied Technology, Qingdao UniversityCollege of Applied Technology, Qingdao UniversityCollege of Applied Technology, Qingdao UniversityCollege of Applied Technology, Qingdao UniversityAbstract Object detection in point clouds is essential for various applications, including autonomous navigation, household robots, and augmented/virtual reality. However, during voxelization and Bird’s Eye View transformation, local point cloud data often remains sparse due to non-target areas and noise points, posing a significant challenge for feature extraction. In this paper, we propose a novel mechanism named Keypoint Guided Sparse Attention (KGSA) to enhance the semantic information of point clouds by calculating Euclidean distances between selected keypoints and others. Additionally, we introduce Instance-wise Box Aligning, a method for expanding predicted boxes and clustering the points within them to achieve precise alignment between predicted bounding boxes and ground-truth targets. Experimental results demonstrate the superiority of our proposed SPBA-Net in 3D object detection on point clouds compared to other state-of-the-art methods.The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.https://doi.org/10.1038/s41598-024-77097-z3D object detectionKeypoint guided sparse attentionInstance-wise box aligning |
| spellingShingle | Haojie Sha Qingrui Gao Hao Zeng Kai Li Wang Li Xuande Zhang Xiaohui Wang SPBA-Net point cloud object detection with sparse attention and box aligning Scientific Reports 3D object detection Keypoint guided sparse attention Instance-wise box aligning |
| title | SPBA-Net point cloud object detection with sparse attention and box aligning |
| title_full | SPBA-Net point cloud object detection with sparse attention and box aligning |
| title_fullStr | SPBA-Net point cloud object detection with sparse attention and box aligning |
| title_full_unstemmed | SPBA-Net point cloud object detection with sparse attention and box aligning |
| title_short | SPBA-Net point cloud object detection with sparse attention and box aligning |
| title_sort | spba net point cloud object detection with sparse attention and box aligning |
| topic | 3D object detection Keypoint guided sparse attention Instance-wise box aligning |
| url | https://doi.org/10.1038/s41598-024-77097-z |
| work_keys_str_mv | AT haojiesha spbanetpointcloudobjectdetectionwithsparseattentionandboxaligning AT qingruigao spbanetpointcloudobjectdetectionwithsparseattentionandboxaligning AT haozeng spbanetpointcloudobjectdetectionwithsparseattentionandboxaligning AT kaili spbanetpointcloudobjectdetectionwithsparseattentionandboxaligning AT wangli spbanetpointcloudobjectdetectionwithsparseattentionandboxaligning AT xuandezhang spbanetpointcloudobjectdetectionwithsparseattentionandboxaligning AT xiaohuiwang spbanetpointcloudobjectdetectionwithsparseattentionandboxaligning |