GA-NET: Global Attention Network for Point Cloud Semantic Segmentation

被引:31
|
作者
Deng, Shuang [1 ,2 ,3 ]
Dong, Qiulei [1 ,2 ,3 ]
机构
[1] Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
[2] Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China
[3] Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Beijing 100190, Peoples R China
基金
中国国家自然科学基金;
关键词
Three-dimensional displays; Feature extraction; Semantics; Computational complexity; Vegetation mapping; Image segmentation; Feeds; 3D point cloud; semantic segmentation; global attention; convolutional neural networks; deep learning;
D O I
10.1109/LSP.2021.3082851
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
How to learn long-range dependencies from 3D point clouds is a challenging problem in 3D point cloud analysis. Addressing this problem, we propose a global attention network for point cloud semantic segmentation, named as GA-Net, consisting of a point-independent global attention module and a point-dependent global attention module for obtaining contextual information of 3D point clouds in this paper. The point-independent global attention module simply shares a global attention map for all 3D points. In the point-dependent global attention module, for each point, a novel random cross attention block using only two randomly sampled subsets is exploited to learn the contextual information of all the points. Additionally, we design a novel point-adaptive aggregation block to replace linear skip connection for aggregating more discriminate features. Extensive experimental results on three 3D public datasets demonstrate that our method outperforms state-of-the-art methods in most cases.
引用
收藏
页码:1300 / 1304
页数:5
相关论文
共 50 条
  • [1] Point attention network for point cloud semantic segmentation
    Dayong REN
    Zhengyi WU
    Jiawei LI
    Piaopiao YU
    Jie GUO
    Mingqiang WEI
    Yanwen GUO
    [J]. Science China(Information Sciences), 2022, (09) : 99 - 112
  • [2] Point attention network for point cloud semantic segmentation
    Ren, Dayong
    Wu, Zhengyi
    Li, Jiawei
    Yu, Piaopiao
    Guo, Jie
    Wei, Mingqiang
    Guo, Yanwen
    [J]. SCIENCE CHINA-INFORMATION SCIENCES, 2022, 65 (09)
  • [3] Point attention network for point cloud semantic segmentation
    Dayong Ren
    Zhengyi Wu
    Jiawei Li
    Piaopiao Yu
    Jie Guo
    Mingqiang Wei
    Yanwen Guo
    [J]. Science China Information Sciences, 2022, 65
  • [4] GA-Net: A geographical attention neural network for the segmentation of body torso tissue composition
    Dai, Jian
    Liu, Tiange
    Torigian, Drew A.
    Tong, Yubing
    Han, Shiwei
    Nie, Pengju
    Zhang, Jing
    Li, Ran
    Xie, Fei
    Udupa, Jayaram K.
    [J]. MEDICAL IMAGE ANALYSIS, 2024, 91
  • [5] PTANet: Triple Attention Network for point cloud semantic segmentation
    Cheng, Haozhe
    Lu, Jian
    Luo, Maoxin
    Liu, Wei
    Zhang, Kaibing
    [J]. ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2021, 102
  • [6] A GLOBAL POINT-SIFT ATTENTION NETWORK FOR 3D POINT CLOUD SEMANTIC SEGMENTATION
    Jia, Meixia
    Li, Aijin
    Wu, Zhaoyang
    [J]. 2019 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2019), 2019, : 5065 - 5068
  • [7] Deep Graph Attention Convolution Network for Point Cloud Semantic Segmentation
    Chai Yujing
    Ma Jie
    Liu Hong
    [J]. LASER & OPTOELECTRONICS PROGRESS, 2021, 58 (12)
  • [8] GA-Net: Ghost convolution adaptive fusion skin lesion segmentation network
    Zhou, Longsong
    Liang, Liming
    Sheng, Xiaoqi
    [J]. COMPUTERS IN BIOLOGY AND MEDICINE, 2023, 164
  • [9] A Dual Attention Neural Network for Airborne LiDAR Point Cloud Semantic Segmentation
    Zhang, Ka
    Ye, Longjie
    Xiao, Wen
    Sheng, Yehua
    Zhang, Shan
    Tao, Xia
    Zhou, Yaqin
    [J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
  • [10] Graph Attention Convolution for Point Cloud Semantic Segmentation
    Wang, Lei
    Huang, Yuchun
    Hou, Yaolin
    Zhang, Shenman
    Shan, Jie
    [J]. 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, : 10288 - 10297