3D point cloud object detection method in view of voxel based on graph convolution network

Networks	Inference time/s	Sensors		Car			Pedestrian			Cyclist
Networks	Inference time/s	LiDAR	Image	Moderate	Hard	Easy	Moderate	Hard	Easy	Moderate	Hard	Easy
MV3D ^[10]	0.36	√	√	62.35%	55.12%	71.09%	N/A	N/A	N/A	N/A	N/A	N/A
MV3D (Lidar)^[10]	0.24	√	×	52.73%	51.31%	66.77%	N/A	N/A	N/A	N/A	N/A	N/A
AVOD ^[11]	0.08	√	√	65.78%	58.38%	73.59%	31.51%	26.98%	38.28%	44.90%	38.80%	60.11%
VoxelNet ^[3]	0.31	√	×	65.46%	62.85%	81.97%	53.42%	48.87%	57.86%	47.65%	45.11%	67.17%
Point-GNN^[6]	0.6	√	×	79.47%	72.29%	88.33%	43.77%	40.14%	51.92%	63.48%	57.08%	78.60%
VGCN (Ours)	0.09	√	×	79.21%	78.58%	89.25%	53.28%	48.74	60.90%	71.82%	68.19%	85.89%

Table 2. Comparison of average precision of BEV object detection on KITTI validation set

View table

View in Article

Table 2. Comparison of average precision of BEV object detection on KITTI validation set

Networks	Inference time/s	Sensors		Car			Pedestrian			Cyclist
Networks	Inference time/s	LiDAR	Image	Moderate	Hard	Easy	Moderate	Hard	Easy	Moderate	Hard	Easy
MV3D ^[10]	0.36	√	√	76.90%	68.49%	86.02%	N/A	N/A	N/A	N/A	N/A	N/A
MV3D (Lidar) ^[10]	0.24	√	×	77.00%	68.94%	85.82%	N/A	N/A	N/A	N/A	N/A	N/A
AVOD ^[11]	0.08	√	√	85.44%	77.73%	86.60%	35.24%	33.97%	42.52%	47.74%	46.55%	63.66%
VoxelNet ^[3]	0.31	√	×	84.81%	78.57%	89.60%	61.05%	56.98%	65.95%	52.18%	50.49%	74.41%
Point-GNN^[6]	0.6	√	×	89.17%	83.90%	93.11%	43.77%	40.14%	51.92%	67.28%	59.67%	81.17%
VGCN (Ours)	0.09	√	×	87.90%	87.33%	90.23%	57.30%	52.72%	64.22%	74.94%	71.57%	87.26%

Table 3. 3D and BEV object detection on KITTI test set

View table

View in Article

Table 3. 3D and BEV object detection on KITTI test set

Benchmarks	Moderate	Easy	Hard
Car(3D)	77.65%	84.47%	73.36%
Car(BEV)	87.16%	90.67%	82.98%
Cyclist(3D)	62.36%	78.47%	55.88%
Cyclist(BEV)	67.04%	81.50%	59.45%
Pedestrian(3D)	37.60%	45.28%	34.96%
Pedestrian(BEV)	42.33%	50.02%	40.05%

Table 4. Effect of the number of graph convolutional layers on vehicle object detection
View table
View in Article
Table 4. Effect of the number of graph convolutional layers on vehicle object detection
Networks Moderate Hard Easy
VoxelNet ^[3] 65.46% 62.85% 81.97%
1-Graph conv 76.32% 75.18% 85.85%
2-Graph convs 79.21% 78.58% 89.25%
4-Graph convs 80.06% 75.72% 86.73%

Tools

Get Citation

Copy Citation Text

Yiqiang Zhao, Akbar Arxidin·, Rui Chen, Yiyao Zhou, Qi Zhang. 3D point cloud object detection method in view of voxel based on graph convolution network[J]. Infrared and Laser Engineering, 2021, 50(10): 20200500

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Image processing

Received: Dec. 20, 2020

Accepted: --

Published Online: Dec. 7, 2021

The Author Email:

DOI:10.3788/IRLA20200500

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. Comparison of average precision of point cloud 3D object detection on KITTI validation set

Table 1. Comparison of average precision of point cloud 3D object detection on KITTI validation set

Table 2. Comparison of average precision of BEV object detection on KITTI validation set

Table 2. Comparison of average precision of BEV object detection on KITTI validation set

Table 3. 3D and BEV object detection on KITTI test set

Table 3. 3D and BEV object detection on KITTI test set

Table 4. Effect of the number of graph convolutional layers on vehicle object detection

Table 4. Effect of the number of graph convolutional layers on vehicle object detection

微信扫一扫：分享