The Graph Neural Network Detector Based on Neighbor Feature Alignment Mechanism in LIDAR Point Clouds
Three-dimensional (3D) object detection has a vital effect on the environmental awareness task of autonomous driving scenarios. At present, the accuracy of 3D object detection has significant improvement potential. In addition, a 3D point cloud is not uniformly distributed on a regular grid because...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Machines |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-1702/11/1/116 |
| _version_ | 1797439482510704640 |
|---|---|
| author | Xinyi Liu Baofeng Zhang Na Liu |
| author_facet | Xinyi Liu Baofeng Zhang Na Liu |
| author_sort | Xinyi Liu |
| collection | DOAJ |
| description | Three-dimensional (3D) object detection has a vital effect on the environmental awareness task of autonomous driving scenarios. At present, the accuracy of 3D object detection has significant improvement potential. In addition, a 3D point cloud is not uniformly distributed on a regular grid because of its disorder, dispersion, and sparseness. The strategy of the convolution neural networks (CNNs) for 3D point cloud feature extraction has the limitations of potential information loss and empty operation. Therefore, we propose a graph neural network (GNN) detector based on neighbor feature alignment mechanism for 3D object detection in LiDAR point clouds. This method exploits the structural information of graphs, and it aggregates the neighbor and edge features to update the state of vertices during the iteration process. This method enables the reduction of the offset error of the vertices, and ensures the invariance of the point cloud in the spatial domain. For experiments performed on the KITTI public benchmark, the results demonstrate that the proposed method achieves competitive experimental results. |
| first_indexed | 2024-03-09T11:53:27Z |
| format | Article |
| id | doaj.art-87357079dda9408bbf5d1abd53148f15 |
| institution | Directory Open Access Journal |
| issn | 2075-1702 |
| language | English |
| last_indexed | 2024-03-09T11:53:27Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Machines |
| spelling | doaj.art-87357079dda9408bbf5d1abd53148f152023-11-30T23:12:07ZengMDPI AGMachines2075-17022023-01-0111111610.3390/machines11010116The Graph Neural Network Detector Based on Neighbor Feature Alignment Mechanism in LIDAR Point CloudsXinyi Liu0Baofeng Zhang1Na Liu2The School of Computer Science and Engineering, Tianjin University of Technology, No. 391 Bin Shui Xi Dao Road, Tianjin 300384, ChinaThe School of Computer Science and Engineering, Tianjin University of Technology, No. 391 Bin Shui Xi Dao Road, Tianjin 300384, ChinaTianjin Key Laboratory for Control Theory and Applications in Complicated System, Tianjin University of Technology, No. 391 Bin Shui Xi Dao Road, Tianjin 300384, ChinaThree-dimensional (3D) object detection has a vital effect on the environmental awareness task of autonomous driving scenarios. At present, the accuracy of 3D object detection has significant improvement potential. In addition, a 3D point cloud is not uniformly distributed on a regular grid because of its disorder, dispersion, and sparseness. The strategy of the convolution neural networks (CNNs) for 3D point cloud feature extraction has the limitations of potential information loss and empty operation. Therefore, we propose a graph neural network (GNN) detector based on neighbor feature alignment mechanism for 3D object detection in LiDAR point clouds. This method exploits the structural information of graphs, and it aggregates the neighbor and edge features to update the state of vertices during the iteration process. This method enables the reduction of the offset error of the vertices, and ensures the invariance of the point cloud in the spatial domain. For experiments performed on the KITTI public benchmark, the results demonstrate that the proposed method achieves competitive experimental results.https://www.mdpi.com/2075-1702/11/1/116LIDAR3D object detectionpoint cloudGNN |
| spellingShingle | Xinyi Liu Baofeng Zhang Na Liu The Graph Neural Network Detector Based on Neighbor Feature Alignment Mechanism in LIDAR Point Clouds Machines LIDAR 3D object detection point cloud GNN |
| title | The Graph Neural Network Detector Based on Neighbor Feature Alignment Mechanism in LIDAR Point Clouds |
| title_full | The Graph Neural Network Detector Based on Neighbor Feature Alignment Mechanism in LIDAR Point Clouds |
| title_fullStr | The Graph Neural Network Detector Based on Neighbor Feature Alignment Mechanism in LIDAR Point Clouds |
| title_full_unstemmed | The Graph Neural Network Detector Based on Neighbor Feature Alignment Mechanism in LIDAR Point Clouds |
| title_short | The Graph Neural Network Detector Based on Neighbor Feature Alignment Mechanism in LIDAR Point Clouds |
| title_sort | graph neural network detector based on neighbor feature alignment mechanism in lidar point clouds |
| topic | LIDAR 3D object detection point cloud GNN |
| url | https://www.mdpi.com/2075-1702/11/1/116 |
| work_keys_str_mv | AT xinyiliu thegraphneuralnetworkdetectorbasedonneighborfeaturealignmentmechanisminlidarpointclouds AT baofengzhang thegraphneuralnetworkdetectorbasedonneighborfeaturealignmentmechanisminlidarpointclouds AT naliu thegraphneuralnetworkdetectorbasedonneighborfeaturealignmentmechanisminlidarpointclouds AT xinyiliu graphneuralnetworkdetectorbasedonneighborfeaturealignmentmechanisminlidarpointclouds AT baofengzhang graphneuralnetworkdetectorbasedonneighborfeaturealignmentmechanisminlidarpointclouds AT naliu graphneuralnetworkdetectorbasedonneighborfeaturealignmentmechanisminlidarpointclouds |