AMFF-Net: An Effective 3D Object Detector Based on Attention and Multi-Scale Feature Fusion
With the advent of autonomous vehicle applications, the importance of LiDAR point cloud 3D object detection cannot be overstated. Recent studies have demonstrated that methods for aggregating features from voxels can accurately and efficiently detect objects in large, complex 3D detection scenes. Ne...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/23/9319 |
| _version_ | 1827592048783392768 |
|---|---|
| author | Guangping Li Zuanfang Mo Bingo Wing-Kuen Ling |
| author_facet | Guangping Li Zuanfang Mo Bingo Wing-Kuen Ling |
| author_sort | Guangping Li |
| collection | DOAJ |
| description | With the advent of autonomous vehicle applications, the importance of LiDAR point cloud 3D object detection cannot be overstated. Recent studies have demonstrated that methods for aggregating features from voxels can accurately and efficiently detect objects in large, complex 3D detection scenes. Nevertheless, most of these methods do not filter background points well and have inferior detection performance for small objects. To ameliorate this issue, this paper proposes an Attention-based and Multiscale Feature Fusion Network (AMFF-Net), which utilizes a Dual-Attention Voxel Feature Extractor (DA-VFE) and a Multi-scale Feature Fusion (MFF) Module to improve the precision and efficiency of 3D object detection. The DA-VFE considers pointwise and channelwise attention and integrates them into the Voxel Feature Extractor (VFE) to enhance key point cloud information in voxels and refine more-representative voxel features. The MFF Module consists of self-calibrated convolutions, a residual structure, and a coordinate attention mechanism, which acts as a 2D Backbone to expand the receptive domain and capture more contextual information, thus better capturing small object locations, enhancing the feature-extraction capability of the network and reducing the computational overhead. We performed evaluations of the proposed model on the nuScenes dataset with a large number of driving scenarios. The experimental results showed that the AMFF-Net achieved 62.8% in the mAP, which significantly boosted the performance of small object detection compared to the baseline network and significantly reduced the computational overhead, while the inference speed remained essentially the same. AMFF-Net also achieved advanced performance on the KITTI dataset. |
| first_indexed | 2024-03-09T01:43:13Z |
| format | Article |
| id | doaj.art-97e60d445acf49d5adf6080ad62f8bfa |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T01:43:13Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-97e60d445acf49d5adf6080ad62f8bfa2023-12-08T15:25:24ZengMDPI AGSensors1424-82202023-11-012323931910.3390/s23239319AMFF-Net: An Effective 3D Object Detector Based on Attention and Multi-Scale Feature FusionGuangping Li0Zuanfang Mo1Bingo Wing-Kuen Ling2School of Information Engineering, Guangdong University of Technology, Guangzhou 510006, ChinaSchool of Information Engineering, Guangdong University of Technology, Guangzhou 510006, ChinaSchool of Information Engineering, Guangdong University of Technology, Guangzhou 510006, ChinaWith the advent of autonomous vehicle applications, the importance of LiDAR point cloud 3D object detection cannot be overstated. Recent studies have demonstrated that methods for aggregating features from voxels can accurately and efficiently detect objects in large, complex 3D detection scenes. Nevertheless, most of these methods do not filter background points well and have inferior detection performance for small objects. To ameliorate this issue, this paper proposes an Attention-based and Multiscale Feature Fusion Network (AMFF-Net), which utilizes a Dual-Attention Voxel Feature Extractor (DA-VFE) and a Multi-scale Feature Fusion (MFF) Module to improve the precision and efficiency of 3D object detection. The DA-VFE considers pointwise and channelwise attention and integrates them into the Voxel Feature Extractor (VFE) to enhance key point cloud information in voxels and refine more-representative voxel features. The MFF Module consists of self-calibrated convolutions, a residual structure, and a coordinate attention mechanism, which acts as a 2D Backbone to expand the receptive domain and capture more contextual information, thus better capturing small object locations, enhancing the feature-extraction capability of the network and reducing the computational overhead. We performed evaluations of the proposed model on the nuScenes dataset with a large number of driving scenarios. The experimental results showed that the AMFF-Net achieved 62.8% in the mAP, which significantly boosted the performance of small object detection compared to the baseline network and significantly reduced the computational overhead, while the inference speed remained essentially the same. AMFF-Net also achieved advanced performance on the KITTI dataset.https://www.mdpi.com/1424-8220/23/23/93193D object detectionLiDARpoint cloudmulti-scale feature fusionattention mechanism |
| spellingShingle | Guangping Li Zuanfang Mo Bingo Wing-Kuen Ling AMFF-Net: An Effective 3D Object Detector Based on Attention and Multi-Scale Feature Fusion Sensors 3D object detection LiDAR point cloud multi-scale feature fusion attention mechanism |
| title | AMFF-Net: An Effective 3D Object Detector Based on Attention and Multi-Scale Feature Fusion |
| title_full | AMFF-Net: An Effective 3D Object Detector Based on Attention and Multi-Scale Feature Fusion |
| title_fullStr | AMFF-Net: An Effective 3D Object Detector Based on Attention and Multi-Scale Feature Fusion |
| title_full_unstemmed | AMFF-Net: An Effective 3D Object Detector Based on Attention and Multi-Scale Feature Fusion |
| title_short | AMFF-Net: An Effective 3D Object Detector Based on Attention and Multi-Scale Feature Fusion |
| title_sort | amff net an effective 3d object detector based on attention and multi scale feature fusion |
| topic | 3D object detection LiDAR point cloud multi-scale feature fusion attention mechanism |
| url | https://www.mdpi.com/1424-8220/23/23/9319 |
| work_keys_str_mv | AT guangpingli amffnetaneffective3dobjectdetectorbasedonattentionandmultiscalefeaturefusion AT zuanfangmo amffnetaneffective3dobjectdetectorbasedonattentionandmultiscalefeaturefusion AT bingowingkuenling amffnetaneffective3dobjectdetectorbasedonattentionandmultiscalefeaturefusion |