计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (9): 249-260.doi: 10.11896/jsjkx.260600097

• 计算机图形学&多媒体 • 上一篇    下一篇

基于对称先验与协同训练多视图子空间聚类的点云部件分割方法

戴叶青, 王春雪, 邓重阳   

  1. 杭州电子科技大学理学院 杭州 310018
  • 收稿日期:2026-04-01 修回日期:2026-07-04 出版日期:2026-09-15 发布日期:2026-09-10
  • 通讯作者: 王春雪(wangcx@hdu.edu.cn)
  • 作者简介:(241070008@hdu.edu.cn)
  • 基金资助:
    浙江省自然科学基金(LQN26F020042);杭州电子科技大学科研启动项目(KYS075624218)

3D Point Cloud Part Segmentation via Symmetry Prior and Co-training Multiview Subspace Clustering

DAI Yeqing, WANG Chunxue, DENG Chongyang   

  1. School of Sciences,Hangzhou Dianzi University,Hangzhou 310018 China
  • Received:2026-04-01 Revised:2026-07-04 Published:2026-09-15 Online:2026-09-10
  • About author:DAI Yeqing,born in 2001,master’s candidate.Her main research interests include computer graphics and graphics processing.
    WANG Chunxue,born in 1988,Ph.D,lecturer,is a member of CCF(No.T9205M).Her main research interests include computer graphics and image processing.
  • Supported by:
    Zhejiang Provincial Natural Science Foundation of China(LQN26F020042) and Hangzhou Dianzi University, China Start-up Fund( KYS075624218).

PDF (PC)

316

摘要： 针对三维点云部件分割中存在的计算复杂度高以及多维特征融合产生的信息冗余等问题,提出一种基于协同训练多视图子空间聚类的点云部件分割方法。首先,以超点过分割为基础,提出一种对称性约束的亲和图优化策略,在精准锁定物体物理结构边界的同时,使后续聚类的计算复杂度大幅下降。其次,面向三维部件的结构特性,为超点构建包含空间位置、鲁棒代表法向以及定制化的多维几何属性的多元特征表达。随后,在协同聚类阶段,利用协同训练多视图子空间聚类框架对核矩阵进行特征分解,以获取低冗余的鲁棒数据表示,并将该表示学习与子空间聚类进行交替优化,实现特征降维与聚类精度的协同演进。最后,采用K近邻多数投票策略对超点标签进行逆向映射与边界平滑优化。在ShapeNet与MeshsegBenchmark-1.0三维点云数据集上的实验结果表明,所提方法有效克服了特征融合带来的冗余干扰,取得了具有竞争力的分割结果;同时,在可视化结果中,该方法在数据集的局部精细结构解析与复杂物理边界保持上,均展现出了高度一致的分割性能。所提方法不仅有效突破了传统多视图聚类应用于大规模点云的算力瓶颈,更结合启发式规则提升了算法的稳健性,为复杂三维物体的结构解析提供了一种可靠的几何驱动途径。

关键词: 点云部件分割, 多视图子空间聚类, 超点过分割, 对称性先验, ShapeNet

Abstract: To address the issues of high computational complexity and information redundancy caused by multi-dimensional feature fusion in 3D point cloud part segmentation,a point cloud part segmentation method based on co-training multi-view subspace clustering is proposed.Firstly,based on superpoint over-segmentation,an affinity graph optimization strategy with symmetry constraints is proposed,which accurately captures the physical structure boundaries of objects while significantly reducing the computational complexity of subsequent clustering.Secondly,targeting the structural characteristics of 3D parts,a multi-feature representation encompassing spatial position,robust representative normals,and customized multi-dimensional geometric attributes is constructed for superpoints.Subsequently,in the co-clustering stage,a co-training multi-view subspace clustering framework is utilized to perform eigendecomposition on the kernel matrix to obtain a robust data representation with low redundancy.This representation learning and the subspace clustering are then alternately optimized to achieve a synergistic evolution of feature dimensionality reduction and clustering accuracy.Finally,a K-nearest neighbor(KNN) majority voting strategy is adopted for the reverse mapping of superpoint labels and boundary smoothing optimization.Experimental results on the ShapeNet and MeshsegBenchmark-1.0 3D point cloud datasets demonstrate that the proposed method effectively overcomes the redundant interference caused by feature fusion and achieves competitive segmentation metrics.Meanwhile,visualization results indicate that the proposed method exhibits highly consistent segmentation performance in parsing local fine structures and preserving complex physical boundaries.The proposed method not only effectively overcomes the computational bottleneck of applying traditional multi-view clustering to large-scale point clouds,but also enhances algorithmic robustness by incorporating heuristic rules,providing a reliable geometry-driven approach for the structural parsing of complex 3D objects.

Key words: Point cloud part segmentation, Multi-view subspace clustering, Superpoint over-segmentation, Symmetry prior, Shape-Net

中图分类号: 

  • TP391.41
[1] RABBANI T,VAN D H.Segmentation of point clouds usingsmoothness constraint [J].International Archives of Photogrammetry,Remote Sensing and Spatial Information Sciences,2006,36(5):248-253.
[2] FISCHLER M A,BOLLES R C.Random Sample Consensus:A paradigm for model fitting withapplications to image analysis and automated cartography [M] //Readings in Computer Vision:Issues,Problems,Principles,and Paradigms.Los Altos:Morgan Kaufmann Publishers,1987:381-395.
[3] ULRIKE V L.A tutorial on spectralclustering [J].Statisticsand Computing,2007,17(4):395-416.
[4] SHI J,MALIK J.Normalized cuts and image segmentation[C] //Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.2026:731-737.
[5] PAPON J,ABRAMOV A,SCHOELER M,et al.Voxel cloudconnectivity segmentation-supervoxels for point clouds [C] //2013 IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2013:2027-2034.
[6] LANDRIEU L,SIMONOVSKY M.Large-scale point cloud semantic segmentation with superpoint graphs [C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR).2018:3298-3307.
[7] 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.
[8] SU H,MAJI S,KALOGERAKIS E,et al.Multi-view convolu-tional neural networks for 3D shape recognition [C] //2015 IEEE International Conference on Computer Vision(ICCV).IEEE,2015:945-953.
[9] BOULCH A,LE SAUX B,AUDEBERT N.Unstructured point cloud semantic labeling using deep segmentation networks [C] //Proceedings of the Workshop on 3D Object Retrieval(3DOR).2017.
[10] KLOKOV R,LEMPITSKY V.Escape from cells:Deep kd-networks for the recognition of 3d point cloud models [C] //Proceedings of the IEEE International Conference on Computer Vision.2017:863-872.
[11] QI C R,SU H,KAICHUN M,et al.PointNet:deep learning on point sets for 3D Classification and segmentation [C] //2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).IEEE,2017:77-85.
[12] QI C R,YI L,SU H,et al.PointNet++:deep hierarchical feature learning on point sets in a metric space [C] //Advances in Neural Information Processing Systems.2017:5099-5108.
[13] THOMAS H,QI C R,DESCHAUD J E,et al.KPConv:flexible and deformable convolution for point clouds [C] //2019 IEEE/CVF International Conference on Computer Vision(ICCV).IEEE,2019:6410-6419.
[14] LI Y,BU R,SUN M,et al.PointCNN:Convolution on X-transformed points [C] //Advances in Neural Information Processing Systems.2018:820-830.
[15] WU W,QI Z,FUXIN L.PointConv:Deep convolutional net-works on 3D point clouds [C] //Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:9621-9630.
[16] YI L,SU H,GUO X,et al.SyncSpecCNN:Synchronized spectral CNN for 3D shape segmentation [C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:2282-2290.
[17] WANG Y,SUN Y,LIU Z,et al.Dynamic graph cnn for learning on point clouds [J].ACM Transactions on Graphics,2019,38(5):1-12.
[18] ZHAO H,JIANG L,JIA J.Point transformer [C] //Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:16259-16268.
[19] ZHOU Y,GU J,CHIANG T Y,et al.Point-SAM:promptable 3D segmentation model for point clouds [C] //International Conference on Learning Representations.2025.
[20] MO K,ZHU S,CHANG A X,et al.PartNet:A large-scalebenchmark for fine-grained and hierarchical part-level 3D object understanding [C] //Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:909-918.
[21] CAO X,ZHANG C,FU H,et al.Diversity-induced multi-view subspace clustering [C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2015:586-594.
[22] ZHANG C,HU Q,FU H,et al.Latent multi-view subspace clustering [C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:4279-4287.
[23] ELHAMIFAR E,VIDAL R.Sparse subspace clustering:algorithm,theory,and applications [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(11):2765-2781.
[24] HU R,FAN L,LIU L.Co-segmentation of 3D shapes via subspace clustering [J].Computer Graphics Forum,2012,31(5):1703-1713.
[25] CHEN Z,YIN K,FISHER M,et al.BAE-NET:Branched au-toencoder for shape co-segmentation [C] //Proceedings of the IEEE/CVF International Conference on Computer Vision.2019:7026-7035.
[26] LIU J,LIU X,YANG Y,et al.Multiview subspace clustering via co-training robust data representation [J].IEEE Transactions on Neural Networks and Learning Systems,2022,33(10):5177-5189.
[27] RUSU R B,COUSINS S.3D is here:point cloud library(PCL) [C] //2011 IEEE International Conference on Robotics and Automation.IEEE,2011:1-4.
[28] HOPPE H,DEROSE T,DUCHAMP T,et al.Surface recon-struction from unorganized points [M] //Seminal Graphics Papers:Pushing the Boundaries,Volume 2.New York:ACM,2023:95-102.
[29] PAULY M,GROSS M,KOBBELT L P.Efficient simplification of point-sampled surfaces [C] //IEEE Visualization,2002.VIS 2002.IEEE,2002:163-170.
[30] ANDREW Y N,MICHAEL I J,YAIR W.On spectral clustering:analysis and an algorithm [C] //Advances in Neural Information Processing Systems.2001:849-856.
[31] SIDI O,VAN KAICK O,KLEIMAN Y,et al.Unsupervised co-segmentation of a set of shapes via descriptor-space spectral clustering [J].ACM Transactions on Graphics,2011,30(6):1-10.
[32] GOLOVINSKIY A,FUNKHOUSER T.Consistent segmenta-tion of 3D models [J].Computers & Graphics,2009,33(3):262-269.
[33] GU S,ZHANG L,ZUO W,et al.Weighted nuclear norm minimization with application to image denoising[C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2014:2862-2869.
[34] LU X,SCHAEFER S,LUO J,et al.Low rank matrix approximation for 3D geometry filtering [J].IEEE Transactions on Visualization and Computer Graphics,2022,28(4):1835-1847.
[35] PATEL V M,VIDAL R.Kernel sparse subspace clustering[C] //2014 IEEE International Conference on Image Proces-sing.IEEE,2014:2849-2853.
[36] SCHÖLKOPF B,SMOLA A J.Learning with kernels:support vector machines,regularization,optimization,and beyond [M].MIT Press,2002.
[37] ZELNIK-MANOR L,PERONA P.Self-tuning spectral clustering [C] //Advances in Neural Information Processing Systems.2004:1601-1608.
[38] CHANG A X,FUNKHOUSER T,GUIBAS L,et al.ShapeNet:An information-rich 3D model repository [J].arXiv:1512.03012,2015.
[39] CHEN X,GOLOVINSKIY A,FUNKHOUSER T.A bench-mark for 3D mesh segmentation [C] //ACM SIGGRAPH 2009 Papers.New York:ACM,2009:1-12.
[40] WU Z,SHOU R,WANG Y,et al.Interactive shape co-segmentation via label propagation [J].Computers & Graphics,2014,38:248-254.
[41] YI L,KIM V G,CEYLAN D,et al.A scalable active framework for region annotation in 3D shape collections [J].ACM Transactions on Graphics,2016,35(6):1-12.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路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