计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (8): 180-187.doi: 10.11896/jsjkx.240900104

• 计算机图形学&多媒体 • 上一篇    下一篇

基于动态平衡和距离抑制的点云语义分割主动学习

曾欣然, 李天瑞, 李崇寿   

  1. 西南交通大学计算机与人工智能学院 成都 611756
  • 收稿日期:2024-09-18 修回日期:2024-11-18 出版日期:2025-08-15 发布日期:2025-08-08
  • 通讯作者: 李崇寿(lics@swjtu.edu.cn)
  • 作者简介:(zxr35@my.swjtu.edu.cn)
  • 基金资助:
    国家自然科学基金(62202395)

Active Learning for Point Cloud Semantic Segmentation Based on Dynamic Balance and DistanceSuppression

ZENG Xinran, LI Tianrui, LI Chongshou   

  1. School of Computing and Artificial Intelligence,Southwest Jiaotong University,Chengdu 611756,China
  • Received:2024-09-18 Revised:2024-11-18 Online:2025-08-15 Published:2025-08-08
  • About author:ZENG Xinran,born in 2000,postgra-duate.Her main research interests include active learning and 3D point cloud semantic segmentation.
    LI Chongshou,born in 1988,Ph.D,associate professor,is a member of CCF(No.J8308M).His main research in-terests include intelligent transportation,data analysis and AI.
  • Supported by:
    National Natural Science Foundation of China(62202395).

PDF (PC)

16

摘要： 近年来,基于深度学习的点云语义分割取得了巨大的成功,但是它在很大程度上依赖于大量密集标注的点云数据。为了降低标注成本,许多弱监督学习方法应运而生。主动学习就是弱监督学习方法的一种,它通过选择点云的一个子集进行标注来降低标注成本。但是,目前的方法对区域信息量的估计没有充分考虑区域内所有点之间的联系,并且之前的多样性选择方法需要耗费较多的时间。为了缓解这些问题,提出一种基于动态平衡和距离抑制的点云语义分割主动学习方法。该方法通过引入区域不一致性来考虑区域中所有点之间的联系,并使用动态平衡策略来调整点级不确定性和区域不一致性的重要性以衡量区域信息量。此外,设计了特征-法线距离抑制策略来选择具有代表性的区域。该策略在考虑区域之间的空间结构时使用了一种更简单的方法,通过删除邻近的相似区域来避免冗余标注,从而提高了多样性选择的效率。在S3DIS和Semantic3D数据集上的实验结果表明,所提框架展现了最先进的性能,并且有效地减少了标注成本和多样性选择时间。

关键词: 点云语义分割, 主动学习, 动态平衡, 点级不确定性, 区域不一致性, 距离抑制

Abstract: In recent years,deep learning-based point cloud semantic segmentation has achieved remarkable success,but it heavily relies on a large amount of densely annotated point cloud data.In order to reduce the annotation cost,many weakly supervised learning methods have emerged,and active learning is one of them.It reduces the annotation cost by selecting a subset of the point cloud for annotation,but the current methods don't fully consider the connection between all points in the region when estimating the amount of regional information,and the previous diversity selection methods take a lot of time.To alleviate these issues,this paper proposes an active learning method for point cloud semantic segmentation based on dynamic balance and distance suppression.The method considers the connection between all points in the region by introducing regional inconsistency,and uses a dynamic balance strategy to adjust the importance of point-level uncertainty and regional inconsistency to measure the amount of regional information.In addition,a feature-normal distance suppression strategy is designed to select representative regions.This strategy uses a simpler method when considering the spatial structure between regions,which avoids redundant labeling by deleting adjacent similar regions,thereby improving the efficiency of diversity selection.Experimental results on the S3DIS and Semantic3D datasets show that the proposed framework demonstrates state-of-the-art performance and effectively reduces the annotation cost and diversity selection time.

Key words: Point cloud semantic segmentation, Active learning, Dynamic balance, Point-level uncertainty, Regional inconsistency, Distance suppression

中图分类号: 

  • TP181
[1]PARK J,KIM C,KIM S,et al.PCSCNet:Fast 3D semantic segmentation of LiDAR point cloud for autonomous car using point convolution and sparse convolution network[J].Expert Systems with Applications,2023,212:118815.
[2]THOMAS H,AGRO B,GRIDSETH M,et al.Self-supervisedlearning of lidar segmentation for autonomous indoor navigation[C]//2021 IEEE International Conference on Robotics and Automation(ICRA).IEEE,2021:14047-14053.
[3]HAN L,ZHENG T,ZHU Y,et al.Live semantic 3d perception for immersive augmented reality[J].IEEE Transactions on Visualization and Computer Graphics,2020,26(5):2012-2022.
[4]QI C R,SU H,MO K,et al.Pointnet:Deep learning on point sets for 3d classification and segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:652-660.
[5]QI C R,YI L,SU H,et al.Pointnet++:Deep hierarchical feature learning on point sets in ametric space[C]//Proceedings of the 31st International Conference onNeural Information Processing Systems.2017:5105-5114.
[6]THOMAS H,QI C R,DESCHAUD J E,et al.Kpconv:Flexible and deformable convolution for point clouds[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2019:6411-6420.
[7]CHOY C,GWAK J Y,SAVARESE S.4D spatio-temporal convnets:Minkowski convolutional neural networks[C]//Procee-dings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:3075-3084.
[8]HU Q,YANG B,XIE L,et al.Randla-net:Efficient semanticsegmentation of large-scale point clouds[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:11108-11117.
[9]TANG H,LIU Z,ZHAO S,et al.Searching efficient 3d architectures with sparse point-voxel convolution[C]//European Conference on Computer Vision.Cham:Springer International Publishing,2020:685-702.
[10]XU Z,YUAN B,ZHAO S,et al.Hierarchical point-based active learning for semi-supervised point cloud semantic segmentation[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2023:18098-18108.
[11]LEI X,GUAN H,MA L,et al.DAAL-WS:A weakly-supervised method integrated with data augmentation and active learning strategies for MLS point cloud semantic segmentation[J].International Journal of Applied Earth Observation and Geoinformation,2024,131:103970.
[12]WU T H,LIU Y C,HUANG Y K,et al.Redal:Region-based and diversity-aware active learning for point cloud semantic segmentation[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:15510-15519.
[13]SHAO F,LUO Y,LIU P,et al.Active learning for point cloud semantic segmentation via spatial-structural diversity reasoning[C]//Proceedings of the 30th ACM International Conference on Multimedia.2022:2575-2585.
[14]GUINARD S,LANDRIEU L.Weakly supervised segmentation-aided classification of urban scenes from 3D LiDAR point clouds[J].The International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2017,42:151-157.
[15]ARMENI I,SENER O,ZAMIR A R,et al.3D semantic parsing of large-scale indoor spaces[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:1534-1543.
[16]HACKEL T,SAVINOV N,LADICKY L,et al.Semantic3d.Net:A new large-scale point cloud classification benchmark[J].ISPRS Annals of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2017,4:91-98.
[17]SU H,MAJI S,KALOGERAKIS E,et al.Multi-view convolutional neural networks for 3d shape recognition[C]//Procee-dings of the IEEE International Conference on Computer Vision.2015:945-953.
[18]BOULCH A,GUERRY J,LE SAUX B,et al.SnapNet:3D point cloud semantic labeling with 2D deep segmentation networks[J].Computers & Graphics,2018,71:189-198.
[19]WU B,WAN A,YUE X,et al.Squeezeseg:Convolutional neural nets with recurrent crf for real-time road-object segmentation from 3d lidar point cloud[C]//2018 IEEE International Confe-rence on Robotics and Automation(ICRA).IEEE,2018:1887-1893.
[20]WU B,ZHOU X,ZHAO S,et al.Squeezesegv2:Improved model structure and unsupervised domain adaptation for road-object segmentation from a lidar point cloud[C]//2019 International Conference on Robotics and Automation(ICRA).IEEE,2019:4376-4382.
[21]MILIOTO A,VIZZO I,BEHLEY J,et al.Rangenet++:Fast and accurate lidar semantic segmentation[C]//2019 IEEE/RSJ International Conference on Intelligent Robots and Systems(IROS).IEEE,2019:4213-4220.
[22]XU C,WU B,WANG Z,et al.Squeezesegv3:Spatially-adaptive convolution for efficient point-cloud segmentation[C]//Compu-ter Vision－ECCV 2020:16th European Conference,Glasgow,UK,Part XXVIII 16.Springer International Publishing,2020:1-19.
[23]ZHAO L,ZHOU H,ZHU X,et al.Lif-seg:Lidar and cameraimage fusion for 3d lidar semantic segmentation[J].IEEE Transactions on Multimedia,2023,26:1158-1168.
[24]MATURANA D,SCHERER S.Voxnet:A 3d convolutionalneural network for real-time object recognition[C]//2015 IEEE/RSJ International Conference on Intelligent Robots and Systems(IROS).IEEE,2015:922-928.
[25]TCHAPMI L,CHOY C,ARMENI I,et al.Segcloud:Semantic segmentation of 3d point clouds[C]//2017 International Confe-rence on 3D Vision(3DV).IEEE,2017:537-547.
[26]GRAHAM B,ENGELCKE M,VAN DER MAATEN L.3d semantic segmentation with submanifold sparse convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:9224-9232.
[27]YAN X,GAO J,LI J,et al.Sparse single sweep lidar point cloud segmentation via learning contextual shape priors from scene completion[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2021:3101-3109.
[28]PARK J,KIM C,KIM S,et al.PCSCNet:Fast 3D semantic segmentation of LiDAR point cloud for autonomous car using point convolution and sparse convolution network[J].Expert Systems with Applications,2023,212:118815.
[29]ZHAO H,JIANG L,FU C W,et al.Pointweb:Enhancing local neighborhood features for point cloud processing[C]//Procee-dings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:5565-5573.
[30]LI Y,BU R,SUN M,et al.PointCNN:Convolution on x-transformed points[C]//Proceedings of the 32nd International Conference on Neural Information Processing Systems.2018:828-838.
[31]WU W,QI Z,LI F.Pointconv:Deep convolutional networks on 3d point clouds[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:9621-9630.
[32]PAPON J,ABRAMOV A,SCHOELER M,et al.Voxel cloud connectivity segmentation-supervoxels for point clouds[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2013:2027-2034.
[33]LI X,WANG X,CHEN X,et al.Unlabeled data selection for active learning in image classification[J].Scientific Reports,2024,14(1):424.
[34]ZHANG S,ZHANG L,LIU Z.Active domain adaptation for semantic segmentation via dynamically balancing domainness and uncertainty[J].Image and Vision Computing,2024,148:105132.
[35]CAI L,XU X,LIEW J H,et al.Revisiting superpixels for active learning in semantic segmentation with realistic annotation costs[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2021:10988-10997.
[36]XU Y,SUN Z,HOEGNER L,et al.Instance segmentation oftrees in urban areas from MLS point clouds using supervoxel contexts and graph-based optimization[C]//2018 10th IAPR Workshop on Pattern Recognition in Remote Sensing(PRRS).IEEE,2018:1-5.
[37]SETTLES B,CRAVEN M.An analysis of active learning strategies for sequence labeling tasks[C]//Proceedings of the 2008 Conference on Empirical Methods in Natural Language Proces-sing.2008:1070-1079.
[38]JOSHI A J,PORIKLI F,PAPANIKOLOPOULOS N.Multi-class active learning for image classification[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2009:2372-2379.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发