AFRNet: Anchor-Free Object Detection Using Roadside LiDAR in Urban Scenes
In urban settings, roadside infrastructure LiDAR is a ground-based remote sensing system that collects 3D sparse point clouds for the traffic object detection of vehicles, pedestrians, and cyclists. Current anchor-free algorithms for 3D point cloud object detection based on roadside infrastructure f...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/16/5/782 |
| _version_ | 1797263933227139072 |
|---|---|
| author | Luyang Wang Jinhui Lan Min Li |
| author_facet | Luyang Wang Jinhui Lan Min Li |
| author_sort | Luyang Wang |
| collection | DOAJ |
| description | In urban settings, roadside infrastructure LiDAR is a ground-based remote sensing system that collects 3D sparse point clouds for the traffic object detection of vehicles, pedestrians, and cyclists. Current anchor-free algorithms for 3D point cloud object detection based on roadside infrastructure face challenges related to inadequate feature extraction, disregard for spatial information in large 3D scenes, and inaccurate object detection. In this study, we propose AFRNet, a two-stage anchor-free detection network, to address the aforementioned challenges. We propose a 3D feature extraction backbone based on the large sparse kernel convolution (LSKC) feature set abstraction module, and incorporate the CBAM attention mechanism to enhance the large scene feature extraction capability and the representation of the point cloud features, enabling the network to prioritize the object of interest. After completing the first stage of center-based prediction, we propose a refinement method based on attentional feature fusion, where fused features incorporating raw point cloud features, voxel features, BEV features, and key point features are used for the second stage of refinement to complete the detection of 3D objects. To evaluate the performance of our detection algorithms, we conducted experiments using roadside LiDAR data from the urban traffic dataset DAIR-V2X, based on the Beijing High-Level Automated Driving Demonstration Area. The experimental results show that AFRNet has an average of 5.27 percent higher detection accuracy than CenterPoint for traffic objects. Comparative tests further confirm that our method achieves high accuracy in roadside LiDAR object detection. |
| first_indexed | 2024-04-25T00:20:52Z |
| format | Article |
| id | doaj.art-333532f41f004ad6b00f6a1bf13336d0 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-04-25T00:20:52Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-333532f41f004ad6b00f6a1bf13336d02024-03-12T16:54:00ZengMDPI AGRemote Sensing2072-42922024-02-0116578210.3390/rs16050782AFRNet: Anchor-Free Object Detection Using Roadside LiDAR in Urban ScenesLuyang Wang0Jinhui Lan1Min Li2Department of Instrument Science and Technology, School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing 100083, ChinaDepartment of Instrument Science and Technology, School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing 100083, ChinaDepartment of Instrument Science and Technology, School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing 100083, ChinaIn urban settings, roadside infrastructure LiDAR is a ground-based remote sensing system that collects 3D sparse point clouds for the traffic object detection of vehicles, pedestrians, and cyclists. Current anchor-free algorithms for 3D point cloud object detection based on roadside infrastructure face challenges related to inadequate feature extraction, disregard for spatial information in large 3D scenes, and inaccurate object detection. In this study, we propose AFRNet, a two-stage anchor-free detection network, to address the aforementioned challenges. We propose a 3D feature extraction backbone based on the large sparse kernel convolution (LSKC) feature set abstraction module, and incorporate the CBAM attention mechanism to enhance the large scene feature extraction capability and the representation of the point cloud features, enabling the network to prioritize the object of interest. After completing the first stage of center-based prediction, we propose a refinement method based on attentional feature fusion, where fused features incorporating raw point cloud features, voxel features, BEV features, and key point features are used for the second stage of refinement to complete the detection of 3D objects. To evaluate the performance of our detection algorithms, we conducted experiments using roadside LiDAR data from the urban traffic dataset DAIR-V2X, based on the Beijing High-Level Automated Driving Demonstration Area. The experimental results show that AFRNet has an average of 5.27 percent higher detection accuracy than CenterPoint for traffic objects. Comparative tests further confirm that our method achieves high accuracy in roadside LiDAR object detection.https://www.mdpi.com/2072-4292/16/5/782object detectionurban areaslaser radarpoint cloudanchor-free network |
| spellingShingle | Luyang Wang Jinhui Lan Min Li AFRNet: Anchor-Free Object Detection Using Roadside LiDAR in Urban Scenes Remote Sensing object detection urban areas laser radar point cloud anchor-free network |
| title | AFRNet: Anchor-Free Object Detection Using Roadside LiDAR in Urban Scenes |
| title_full | AFRNet: Anchor-Free Object Detection Using Roadside LiDAR in Urban Scenes |
| title_fullStr | AFRNet: Anchor-Free Object Detection Using Roadside LiDAR in Urban Scenes |
| title_full_unstemmed | AFRNet: Anchor-Free Object Detection Using Roadside LiDAR in Urban Scenes |
| title_short | AFRNet: Anchor-Free Object Detection Using Roadside LiDAR in Urban Scenes |
| title_sort | afrnet anchor free object detection using roadside lidar in urban scenes |
| topic | object detection urban areas laser radar point cloud anchor-free network |
| url | https://www.mdpi.com/2072-4292/16/5/782 |
| work_keys_str_mv | AT luyangwang afrnetanchorfreeobjectdetectionusingroadsidelidarinurbanscenes AT jinhuilan afrnetanchorfreeobjectdetectionusingroadsidelidarinurbanscenes AT minli afrnetanchorfreeobjectdetectionusingroadsidelidarinurbanscenes |