IDA-Net: Intensity-distribution aware networks for semantic segmentation of 3D MLS point clouds in indoor corridor environments

被引:3
作者
Luo, Zhipeng [1 ,2 ]
Chen, Pengxin [1 ,2 ]
Shi, Wenzhong [1 ,2 ]
Li, Jonathan [3 ,4 ]
机构
[1] Hong Kong Polytech Univ, Dept Land Surveying & Geoinformat, Hong Kong 999077, Peoples R China
[2] Hong Kong Polytech Univ, Otto Poon Charitable Fdn, Smart Cities Res Inst, Hong Kong 999077, Peoples R China
[3] Univ Waterloo, Dept Geog & Environm Management, Waterloo, ON, Canada
[4] Univ Waterloo, Dept Syst Design Engn, Waterloo, ON, Canada
关键词
3D MLS point cloud; Semantic segmentation; Intensity-distribution aware; Two-stage embedding network; OBJECT RECOGNITION; SURFACE; REPRESENTATION; CLASSIFICATION; FEATURES;
D O I
10.1016/j.jag.2022.102904
中图分类号
TP7 [遥感技术];
学科分类号
081102 ; 0816 ; 081602 ; 083002 ; 1404 ;
摘要
Semantic segmentation of 3D mobile laser scanning point clouds is the foundational task for scene understanding in several fields. Most existing segmentation methods tend to simply stack the common point attributes, such as the coordinates and intensity, but ignore their heterogeneous. This paper presents IDA-Net, an intensity-distribution aware network that mines the uniqueness and discrepancy of these two modalities in a separate way for point cloud segmentation under indoor corridor environments. Specifically, IDA-Net consists of two key components. Firstly, an intensity-distribution aware (IDA) descriptor is proposed to mine the intensity distri-bution pattern. It outputs a multi-channel mask for each point to represent the intensity distribution information. Secondly, a two-stage embedding network is designed to fuse the coordinates and intensity information effi-ciently. It includes a guiding operation in training stage and a refining operation in testing stage. IDA-Net was evaluated on two indoor corridor areas. Experimental results show that the proposed method significantly im-proves the performance of segmentation. Specifically, with backbone of KPConv, IDA-Net achieves high mIoU of 90.58% and 88.94% on the above two testing areas respectively, which demonstrates the superiority of the designed IDA descriptor and two-stage embedding network.
引用
收藏
页数:12
相关论文
共 52 条
  • [1] [Anonymous], 2017, P 3DOR EUR
  • [2] ConvPoint: Continuous convolutions for point cloud processing
    Boulch, Alexandre
    [J]. COMPUTERS & GRAPHICS-UK, 2020, 88 : 24 - 34
  • [3] Extensive Tests of Autonomous Driving Technologies
    Broggi, Alberto
    Buzzoni, Michele
    Debattisti, Stefano
    Grisleri, Paolo
    Laghi, Maria Chiara
    Medici, Paolo
    Versari, Pietro
    [J]. IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2013, 14 (03) : 1403 - 1415
  • [4] 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks
    Choy, Christopher
    Gwak, JunYoung
    Savarese, Silvio
    [J]. 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, : 3070 - 3079
  • [5] Shape Completion using 3D-Encoder-Predictor CNNs and Shape Synthesis
    Dai, Angela
    Qi, Charles Ruizhongtai
    Niessner, Matthias
    [J]. 30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, : 6545 - 6554
  • [6] Know What Your Neighbors Do: 3D Semantic Segmentation of Point Clouds
    Engelmann, Francis
    Kontogianni, Theodora
    Schult, Jonas
    Leibe, Bastian
    [J]. COMPUTER VISION - ECCV 2018 WORKSHOPS, PT III, 2019, 11131 : 395 - 409
  • [7] On-Board Object Detection: Multicue, Multimodal, and Multiview Random Forest of Local Experts
    Gonzalez, Alejandro
    Vazquez, David
    Lopez, Antonio M.
    Amores, Jaume
    [J]. IEEE TRANSACTIONS ON CYBERNETICS, 2017, 47 (11) : 3980 - 3990
  • [8] 3D Semantic Segmentation with Submanifold Sparse Convolutional Networks
    Graham, Benjamin
    Engelcke, Martin
    van der Maaten, Laurens
    [J]. 2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, : 9224 - 9232
  • [9] Deep Learning for 3D Point Clouds: A Survey
    Guo, Yulan
    Wang, Hanyun
    Hu, Qingyong
    Liu, Hao
    Liu, Li
    Bennamoun, Mohammed
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2021, 43 (12) : 4338 - 4364
  • [10] A novel local surface feature for 3D object recognition under clutter and occlusion
    Guo, Yulan
    Sohel, Ferdous
    Bennamoun, Mohammed
    Wan, Jianwei
    Lu, Min
    [J]. INFORMATION SCIENCES, 2015, 293 : 196 - 213