Performance of Graph Neural Networks for Point Cloud Applications

被引:0
|
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
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
  • [1] GRAPH ATTENTION NEURAL NETWORKS FOR POINT CLOUD RECOGNITION
    Li, Zongmin
    Zhang, Jun
    Li, Guanlin
    Liu, Yujie
    Li, Siyuan
    2019 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO (ICME), 2019, : 387 - 392
  • [2] POINT CLOUD NORMAL ESTIMATION WITH GRAPH-CONVOLUTIONAL NEURAL NETWORKS
    Pistilli, Francesca
    Fracastoro, Giulia
    Valsesia, Diego
    Magli, Enrico
    2020 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO WORKSHOPS (ICMEW), 2020,
  • [3] Cloud Native Applications Profiling using a Graph Neural Networks Approach
    Boukhtouta, Amine
    Madi, Taous
    Pourzandi, Makan
    Alameddine, Hyame A.
    2022 IEEE FUTURE NETWORKS WORLD FORUM, FNWF, 2022, : 220 - 227
  • [4] A Survey on Graph Neural Networks for Microservice-Based Cloud Applications
    Nguyen, Hoa Xuan
    Zhu, Shaoshu
    Liu, Mingming
    SENSORS, 2022, 22 (23)
  • [5] Towards Efficient Point Cloud Graph Neural Networks Through Architectural Simplification
    Tailor, Shyam A.
    de Jong, Rene
    Azevedo, Tiago
    Mattina, Matthew
    Maji, Partha
    2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW 2021), 2021, : 2095 - 2104
  • [6] HPGNN: Using Hierarchical Graph Neural Networks for Outdoor Point Cloud Processing
    Thieshanthan, Arulmolivarman
    Niwarthana, Amashi
    Somarathne, Pamuditha
    Wickremasinghe, Tharindu
    Rodrigo, Ranga
    2022 26TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2022, : 2700 - 2706
  • [7] Geometric attentional dynamic graph convolutional neural networks for point cloud analysis
    Cui, Yiming
    Liu, Xin
    Liu, Hongmin
    Zhang, Jiyong
    Zare, Alina
    Fan, Bin
    NEUROCOMPUTING, 2021, 432 : 300 - 310
  • [8] Graph neural networks for construction applications
    Jia, Yilong
    Wang, Jun
    Shou, Wenchi
    Hosseini, M. Reza
    Bai, Yu
    AUTOMATION IN CONSTRUCTION, 2023, 154
  • [9] Survey of Graph Neural Networks and Applications
    Liang, Fan
    Qian, Cheng
    Yu, Wei
    Griffith, David
    Golmie, Nada
    WIRELESS COMMUNICATIONS & MOBILE COMPUTING, 2022, 2022
  • [10] Research on Three-Dimensional Point Cloud Reconstruction Method Based on Graph Neural Networks
    Ma, Ruocheng
    Gao, Xiang
    Song, Zhaoxiang
    2024 3RD INTERNATIONAL CONFERENCE ON IMAGE PROCESSING AND MEDIA COMPUTING, ICIPMC 2024, 2024, : 106 - 113