Performance of Graph Neural Networks for Point Cloud Applications

被引:1
|
作者
Parikh, Dhruv [1 ]
Zhang, Bingyi [1 ]
Kannan, Rajgopal [2 ]
Prasanna, Viktor [1 ]
Busart, Carl [2 ]
机构
[1] Univ Southern Calif, Los Angeles, CA 90007 USA
[2] DEVCOM US Army Res Lab, Adelphi, MD USA
来源
2023 IEEE HIGH PERFORMANCE EXTREME COMPUTING CONFERENCE, HPEC | 2023年
关键词
Graph neural network; point cloud; k-nearest neighbors; dynamic graph construction; performance profiling;
D O I
10.1109/HPEC58863.2023.10363595
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Graph Neural Networks (GNNs) have gained significant momentum recently due to their capability to learn on unstructured graph data. Dynamic GNNs (DGNNs) are the current state-of-the-art for point cloud applications; such applications (viz. autonomous driving) require real-time processing at the edge with tight latency and memory constraints. Conducting performance analysis on such DGNNs, thus, becomes a crucial task to evaluate network suitability. This paper presents a profiling analysis of EdgeConv-based DGNNs applied to point cloud inputs. We assess their inference performance in terms of end-to-end latency and memory consumption on state-of-the-art CPU and GPU platforms. The EdgeConv layer has two stages: (1) dynamic graph generation using k-Nearest Neighbors (kNN) and, (2) node feature updation. The addition of dynamic graph generation via kNN in each (EdgeConv) layer enhances network performance compared to networks that work with the same static graph in each layer; such performance enhancement comes, however, at the added computational cost associated with the dynamic graph generation stage (via kNN algorithm). Understanding its costs is essential for identifying the performance bottleneck and exploring potential avenues for hardware acceleration. To this end, this paper aims to shed light on the performance characteristics of EdgeConv-based DGNNs for point cloud inputs. Our performance analysis on a state-of-the-art EdgeConv network for classification shows that the dynamic graph construction via kNN takes up upwards of 95% of network latency on the GPU and almost 90% on the CPU. Moreover, we propose a quasi-Dynamic Graph Neural Network (qDGNN) that halts dynamic graph updates after a specific depth within the network to significantly reduce the latency on both CPU and GPU whilst matching the original networks inference accuracy.
引用
收藏
页数:7
相关论文
共 50 条
  • [21] A Graph-Voxel Joint Convolution Neural Network for ALS Point Cloud Segmentation
    Zhang, Jinming
    Hu, Xiangyun
    Dai, Hengming
    IEEE ACCESS, 2020, 8 : 139781 - 139791
  • [22] GRAPH NEURAL NETWORK WITH MULTI-KERNEL LEARNING FOR MULTISPECTRAL POINT CLOUD CLASSIFICATION
    Zhang, Zifeng
    Wang, Qingwang
    Wang, Mingye
    Shen, Tao
    IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, 2023, : 970 - 973
  • [23] An Efficient Accelerator for Point-based and Voxel-based Point Cloud Neural Networks
    Yang, Xinhao
    Fu, Tianyu
    Dai, Guohao
    Zeng, Shulin
    Zhong, Kai
    Hong, Ke
    Wang, Yu
    2023 60TH ACM/IEEE DESIGN AUTOMATION CONFERENCE, DAC, 2023,
  • [24] Analyzing point cloud of coal mining process in much dust environment based on dynamic graph convolution neural network
    Zhizhong Xing
    Shuanfeng Zhao
    Wei Guo
    Xiaojun Guo
    Shenquan Wang
    Mingyue Li
    Yuan Wang
    Haitao He
    Environmental Science and Pollution Research, 2023, 30 : 4044 - 4061
  • [25] Analyzing point cloud of coal mining process in much dust environment based on dynamic graph convolution neural network
    Xing, Zhizhong
    Zhao, Shuanfeng
    Guo, Wei
    Guo, Xiaojun
    Wang, Shenquan
    Li, Mingyue
    Wang, Yuan
    He, Haitao
    ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH, 2023, 30 (02) : 4044 - 4061
  • [26] Wheel Center Detection from Point Cloud Based on Recurrent Neural Networks
    Zheng, Wenyuan
    Mei, Tiancan
    Zeng, Zhiwen
    Sui, Libin
    Ji, Hong
    2018 3RD IEEE INTERNATIONAL CONFERENCE ON ADVANCED ROBOTICS AND MECHATRONICS (IEEE ICARM), 2018, : 594 - 599
  • [27] Can point-cloud based neural networks learn fingerprint variability?
    Sollinger, Dominik
    Jochl, Robert
    Kirchgasser, Simon
    Uhl, Andreas
    PROCEEDINGS OF THE 21ST 2022 INTERNATIONAL CONFERENCE OF THE BIOMETRICS SPECIAL INTEREST GROUP (BIOSIG 2022), 2022, P-329
  • [28] 3-D Point Cloud Registration Using Convolutional Neural Networks
    Chang, Wen-Chung
    Van-Toan Pham
    APPLIED SCIENCES-BASEL, 2019, 9 (16):
  • [29] Two-Stage Point Cloud Registration Framework Based on Graph Neural Network and Attention
    Zhang, Xiaoqian
    Li, Junlin
    Zhang, Wei
    Xu, Yansong
    Li, Feng
    ELECTRONICS, 2024, 13 (03)
  • [30] CVA-GNN: Convolutional Vicinity Aggregation Graph Neural Network for Point Cloud Classification
    Jakub, Walczak
    Patryk, Najgebauer
    Rafal, Scherer
    Wojciechowski, Adam
    2021 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2021,