Feature fusion network based on attention mechanism for 3D semantic segmentation of point clouds

被引:21
|
作者
Zhou, Heng [1 ]
Fang, Zhijun [1 ]
Gao, Yongbin [1 ]
Huang, Bo [1 ]
Zhong, Cengsi [1 ]
Shang, Ruoxi [2 ]
机构
[1] Shanghai Univ Engn Sci, 333 Longteng Rd, Shanghai 201620, Peoples R China
[2] Univ Calif Berkeley, Berkeley, CA 94720 USA
基金
中国国家自然科学基金;
关键词
3D Semantic segmentation; Point clouds; Feature fusion; Attention mechanism;
D O I
10.1016/j.patrec.2020.03.021
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
3D scene parsing has always been a hot topic and point clouds are efficient data format to represent scenes. The semantic segmentation of point clouds is critical to the 3D scene, which is a challenging problem due to the unordered structure of point clouds. The max-pooling operation is typically used to obtain the order invariant features, while the point-wise features are destroyed after the max-pooling operation. In this paper, we propose a feature fusion network that fuses point-wise features and local features by attention mechanism to compensate for the loss caused by max-pooling operation. By incorporating point-wise features into local features, the point-wise variation is preserved to obtain a refined segmentation accuracy, and the attention mechanism is used to measure the importance of the pointwise features and local features for each 3D point. Extensive experiments show that our method achieves better performances than other prestigious methods. (C) 2020 Published by Elsevier B.V.
引用
收藏
页码:327 / 333
页数:7
相关论文
共 50 条
  • [1] Point attention network for semantic segmentation of 3D point clouds
    Feng, Mingtao
    Zhang, Liang
    Lin, Xuefei
    Gilani, Syed Zulqarnain
    Mian, Ajmal
    [J]. PATTERN RECOGNITION, 2020, 107
  • [2] Semantic segmentation of 3D point cloud based on self-attention feature fusion group convolutional neural network
    Yang, Jun
    Li, Bozan
    [J]. Guangxue Jingmi Gongcheng/Optics and Precision Engineering, 2022, 30 (07): : 840 - 853
  • [3] DPRNet: Deep 3D Point Based Residual Network for Semantic Segmentation and Classification of 3D Point Clouds
    Arshad, Saira
    Shahzad, Muhammad
    Riaz, Qaiser
    Fraz, Muhammad Moazam
    [J]. IEEE ACCESS, 2019, 7 : 68892 - 68904
  • [4] Attention-Based Joint Semantic-Instance Segmentation of 3D Point Clouds
    Hao, Wen
    Wang, Hongxiao
    Liang, Wei
    Zhao, Minghua
    Xia, Zhaolin
    [J]. ADVANCES IN ELECTRICAL AND COMPUTER ENGINEERING, 2022, 22 (02) : 19 - 28
  • [5] Point cloud semantic segmentation based on local feature fusion and multilayer attention network
    Wen, Junjie
    Ma, Jie
    Zhao, Yuehua
    Nie, Tong
    Sun, Mengxuan
    Fan, Ziming
    [J]. IET COMPUTER VISION, 2024, 18 (03) : 381 - 392
  • [6] PointFusionNet: Point feature fusion network for 3D point clouds analysis
    Liang, Pan
    Fang, Zhijun
    Huang, Bo
    Zhou, Heng
    Tang, Xianhua
    Zhong, Cengsi
    [J]. APPLIED INTELLIGENCE, 2021, 51 (04) : 2063 - 2076
  • [7] PointFusionNet: Point feature fusion network for 3D point clouds analysis
    Pan Liang
    Zhijun Fang
    Bo Huang
    Heng Zhou
    Xianhua Tang
    Cengsi Zhong
    [J]. Applied Intelligence, 2021, 51 : 2063 - 2076
  • [8] FGCN: Deep Feature-based Graph Convolutional Network for Semantic Segmentation of Urban 3D Point Clouds
    Khan, Saqib Ali
    Shi, Yilei
    Shahzad, Muhammad
    Zhu, Xiao Xiang
    [J]. 2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW 2020), 2020, : 778 - 787
  • [9] Local Fusion Attention Network for Semantic Segmentation of Building Facade Point Clouds
    Su, Yanfei
    Liu, Weiquan
    Cheng, Ming
    Yuan, Zhimin
    Wang, Cheng
    [J]. IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2022, 19
  • [10] Lightweight Semantic Segmentation Network based on Attention Feature Fusion
    Kuang, Xianyan
    Liu, Ping
    Chen, Yixi
    Zhang, Jianhua
    [J]. ENGINEERING LETTERS, 2023, 31 (04)