结合图采样和图注意力的3D目标检测方法

doi:10.3778/j.issn.1002-8331.2202-0075

摘要/Abstract

摘要： 在点云中进行三维目标检测时，小目标和复杂背景下目标的检测精度不足是突出的问题之一。针对该问题，提出了一种基于图采样和图注意力机制的3D点云目标检测方法。减小基准网络下采样体素大小以保持小目标的点云密度；引入图采样降低在点云中构造拓扑图的代价；通过对图采样前后的图分别嵌入自注意力机制，提高网络的特征提取能力。在KITTI数据集上与基准网络Point-GNN相比，对汽车目标在复杂场景上的检测精度提升了1.96%，对行人与骑行者目标在中等难度场景和复杂场景上的检测精度分别提升4.21%和2.57%；与Point-GNN相比，减少了15%的训练时间。实验结果表明，设计的方法对于3D点云中小目标和复杂背景下目标的检测更加有效，图采样方法还能够提升模型的训练效率。

关键词: 点云, 3D目标检测, 图神经网络, 图采样, 图注意力机制

Abstract: For the task of 3D object detection in point clouds, there are objects have a small scale or appear in complex scenes, which makes them have a lower detection accuracy. Therefore, a 3D object detection method based on graph sampling and graph attention mechanism using the point clouds is proposed. Firstly, the method reduces the size of down-sampling voxels to maintain the point clouds density of small objects, and then introduces graph sampling technology to reduce the cost of constructing topological graphs in the point clouds for feature extraction. Finally, the embedded self-attention mechanism in graphs before and after the graph sampling is used to enhance the feature extraction ability of the network. Compared with the benchmark on the KITTI dataset, proposed method improves the detection accuracy of car in hard scenes by 1.96%, and improves the detection accuracy of pedestrian and cyclist in moderate scenes and hard scenes with 4.21% and 2.57% respectively. Besides, the training time of proposed method is reduced by 15%. These demonstrate superior performance in detection accuracy of small objects in point clouds and the sampling method can improve the training efficiency of the model.

Key words: point cloud, 3D object detection, graph neural network, graph sampling, graph attention mechanism

李文举, 储王慧, 崔柳, 苏攀, 张干. 结合图采样和图注意力的3D目标检测方法[J]. 计算机工程与应用, 2023, 59(9): 237-244.

LI Wenju, CHU Wanghui, CUI Liu, SU Pan, ZHANG Gan. 3D Object Detection Method Combining on Graph Sampling and Graph Attention[J]. Computer Engineering and Applications, 2023, 59(9): 237-244.

参考文献

[1] 夏显召，朱世贤，周意遥，等.基于阈值的激光雷达K均值聚类算法[J].北京航空航天大学学报，2020，46（1）：115-121.
XIA X Z，ZHU S X，ZHOU Y Y，el al.LiDAR K-means clustering algorithm based on threshold[J].Journal of Beijing University of Aeronautics and Astronautics，2020，46（1）：115-121.
[2] 孙一珺，胡辉，李子钥，等.适用于点云数据的注意力机制研究[J].计算机工程与应用，2022，58（23）：254-260.
SUN Y J，HU H，LI Z Y，et al.Research on attention mechanism for point cloud data[J].Computer Engineering and Applications，2022，58（23）：254-260.
[3] SHI S，WANG X，LI H.Pointrcnn：3d object proposal generation and detection from point cloud[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2019：770-779.
[4] QI C R，YI L，SU H，et al.PointNet++：deep hierarchical feature learning on point sets in a metric space[C]//Advances in Neural Information Processing Systems，2017.
[5] ZHOU Y，TUZEL O.Voxelnet：end-to-end learning for point cloud based 3d object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2018：4490-4499.
[6] TU J，WANG P，LIU F.PP-RCNN：point-pillars feature set abstraction for 3D real-time object detection[C]//2021 International Joint Conference on Neural Networks（IJCNN），2021：1-8.
[7] KIPF T N，WELLING M.Semi-supervised classification with graph convolutional networks[J].arXiv：1609.02907，2016.
[8] 赵毅强，艾西丁·艾克白尔，陈瑞，等.基于体素化图卷积网络的三维点云目标检测方法[J].红外与激光工程，2021，50（10）：281-289.
ZHAO Y Q，ARXIDIN·AKBAR，CHEN R，et al.3D point cloud object detection method in view of voxel based on graph convolution network[J].Infrared and Laser Engineering，2021，50（10）：281-289.
[9] SHI W J，RAGUNATHAN R.Point-GNN：graph neural network for 3D object detection in a point cloud[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，Seattle，WA，USA，2020：1711-1719.
[10] LIU Z，ZHAO X，HUANG T，et al.Tanet：robust 3d object detection from point clouds with triple attention[C]//Proceedings of the AAAI Conference on Artificial Intelligence，2020：11677-11684.
[11] LI B.3d fully convolutional network for vehicle detection in point cloud[C]//2017 IEEE/RSJ International Conference on Intelligent Robots and Systems（IROS），2017：1513-1518.
[12] YANG B，LUO W，URTASUN R.Pixor：real-time 3d object detection from point clouds[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2018：7652-7660.
[13] KUANG H，WANG B，AN J，et al.Voxel-FPN：multi-scale voxel feature aggregation for 3D object detection from LIDAR point clouds[J].Sensors，2020，20（3）：704.
[14] HE C，ZENG H，HUANG J，et al.Structure aware single-stage 3d object detection from point cloud[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2020：11873-11882.
[15] CHEN X，MA H，WAN J，et al.Multi-view 3d object detection network for autonomous driving[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2017：1907-1915.
[16] HE Q，WANG Z，ZENG H，et al.SVGA-net：sparse voxel-graph attention network for 3D object detection from point clouds[J].arXiv：2006.04043，2020.
[17] PAIGWAR A，SIERRA-GONZALEZ D，ERKENT O，et al.Frustum-pointpillars：a multi-stage approach for 3d object detection using rgb camera and lidar[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision，2021：2926-2933.
[18] ZENG H，ZHOU H，SRIVASTAVA A，et al.GraphSAINT：graph sampling based inductive learning method[C]//International Conference on Learning Representations，2020.
[19] WANG Y，HAO Q.Towards more robust GNN training with graph normalization for GraphSAINT[C]//International Conference on Applied Informatics.Cham：Springer，2021：82-93.
[20] VELICKOVIC P，CUCURULL G，CASANOVA A，et al.Graph attention networks[J].Stat，2018，1050：4.
[21] GEIGER A，LENZ P，STILLER C，et al.Vision meets robotics：lhe kitti dataset[J].The International Journal of Robotics Research，2013，32（11）：1231-1237.