| 204 | 0 | 25 |
| 下载次数 | 被引频次 | 阅读次数 |
点云在目标检测等领域中有很重要的作用。目前对点云特征提取之前需要对点云进行预处理，通常的处理方法包括将点云投影到二维平面、利用栅格法将点云体素化或直接基于原始点云进行处理。这些方法都会损失一定的原始信息，直接处理点云的原始信息面临计算量很大的问题。为解决上述问题，设计了一个平行图卷积神经网络。在将点云转化为图的基础上，利用图卷积神经网络不同的卷积核来提取点云的特征，并通过提出的一种注意力机制进行不同层次的特征融合。提出的方法可以在最大程度上保留点云原始信息的基础上，从不同的角度提取图的特征。在KITTI数据集上的实验表明，所提出的方法是有效的，并且取得了较好的结果。
Abstract：Point clouds play an important role in fields such as target detection. Currently, point clouds usually need to be pre-processed before feature extraction. The usual processing methods include projecting the point cloud onto a 2D plane, voxelizing the point cloud using the raster method, or processing the point cloud directly based on the original point cloud. However, all these methods lose some original information, and the direct processing of original information of the point cloud faces the problem of large computational volume. To solve the above problems, a parallel graph convolutional neural network is designed. On the basis of converting point clouds into graphs, different convolutional kernels of the graph convolutional neural network are used to extract the features of point clouds. And an attention mechanism is proposed to perform feature fusion at different levels. The proposed method extracts the features of the graph from different perspectives while preserving the original information of the point cloud to the maximum extent. Experiments on the KITTI dataset show that the proposed method is effective and achieves better results.
[1] GEIGER A,LENZ P,URTASUN R.Are We Ready for Autonomous Driving?The KITTI Vision Benchmark Suite[C]//2012 IEEE Conference on Computer Vision and Pattern Recognition.Providence:IEEE,2012:3354-3361.
[2] PARK Y,LEPETIT V,WOO W.Multiple 3D Object Tracking for Augmented Reality[C]//2008 7th IEEE/ACM International Symposium on Mixed and Augmented Reality.Cambridge:IEEE,2008:117-120.
[3] NIE Y Y,HAN X G,NIEβNER M.RfD-Net:Point Scene Understanding by Semantic Instance Reconstruction[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition.[S.l.]:IEEE,2021:4608-4618.
[4] DEFFERRARD M,BRESSON X,VANDERGHEYNST P.Convolutional Neural Networks on Graphs with Fast Loca-lized Spectral Filtering[C]//30th International Conference on Neural Information Processing Systems.Barcelona:Curran Associates Inc.,2016:3844-3852.
[5] HAMILTON W L,REX Y,LESKOVEC J.Inductive Representation Learning on Large Graphs[C]//31st International Conference on Neural Information Processing Systems.Long Beach:Curran Associates Inc.,2017:1025-1035.
[6] VELICKOVIC P,CUCURULL G,CASANOVA A,et al.Graph Attention Networks[J/OL].(2018-02-04) [2022-08-26].https://arxiv.org/abs/1710.10903.
[7] CHEN X,MA H,WAN J,et al.Multi-view 3D Object Detection Network for Autonomous Driving[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Honolulu:IEEE,2017:1907-1915,
[8] YAN Y,MAO Y,LI B.SECOND:Sparsely Embedded Convolutional Detection[J].Sensors,2018,18(10):3337.
[9] QI C R,YI L,SU H,et al.PointNet++:Deep Hierarchical Feature Learning on Point Sets in a Metric Space[J/OL].(2017-06-07) [2022-08-26].https://arxiv.org/abs/1706.02413.
[10] YANG Z,SUN Y,LIU S,et al.STD:Sparse-to-dense 3D Object Detector for Point Cloud[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV).Seoul:IEEE,2019:1951-1960.
[11] LANG A H,VORA S,CAESAR H,et al.PointPillars:Fast Encoders for Object Detection From Point Clouds[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Long Beach:IEEE,2019:12689-12697.
[12] LANDRIEU L,SIMONOVSKY M.Large-scale Point Cloud Semantic Segmentation with Super-point Graphs[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City:IEEE,2018:4558-4567.
[13] SHI W,RAJKUMAR R.Point-GNN:Graph Neural Network for 3D Object Detection in a Point Cloud[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Seattle:IEEE,2020:1708-1716.
[14] HE Q D,WANG Z N,ZENG H,et al.SVGA-Net:Sparse Voxel-graph Attention Network for 3D Object Detection from Point Clouds[C]//36th AAAI Conference on Artificial Intelligence.[S.l.]:AAAI,2022:870-878.
[15] XIE S,LIU S,CHEN Z,et al.Attentional ShapeContextNet for Point Cloud Recognition[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City:IEEE,2018:4606-4615.
[16] ZHOU Y,TUZEL O.VoxelNet:End-to-End Learning for Point Cloud Based 3D Object Detection[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City:IEEE,2018:4490-4499.
[17] SHI S,WANG X,LI H.PointRCNN:3D Object Proposal Generation and Detection from Point Cloud[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Long Beach:IEEE,2019:770-779.
[18] LIANG M,YANG B,CHEN Y,et al.Multi-task Multi-sensor Fusion for 3D Object Detection[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Long Beach:IEEE,2019:7345-7353.
基本信息:
DOI：
中图分类号:TP391.41;TP183
引用信息:
[1]刘书勇,付轩硕,李超.基于并行图神经网络的3D点云目标检测[J].无线电工程,2023,53(07):1686-1692.
基金信息:
黑龙江省自然科学基金(JJ2019LH2160)~~