三维点云孔洞修复方法综述Review of three-dimensional point cloud completion methods
赵江洪,孙铭悦,王殷瑞,窦新铜,张晓光
摘要(Abstract)：
针对点云数据缺失问题,该文综合国内外大量点云修复技术研究。三维模型构建在自动驾驶、逆向工程领域中发挥越来越大的作用,三维点云数据是其中的重要数据源。利用三维激光扫描设备,可以高效、准确、实时的获取被测物体表面三维空间坐标。但是由于模型物体遮挡或者环境等原因,不可避免的会出现点云缺失的状况,这会对物体三维重建等后续处理造成一定的影响。然而,在三维点云孔洞修复方面还缺少比较系统完善的综述。本文从基于几何、基于模型检索、基于深度学习3个方面对当前主流的对修复技术进行了综合分析。文章对3种修复方法进行了概括,总结现有各种技术修复方法的优劣,同时展望了未来的发展趋势。
关键词(KeyWords)： 点云缺失;点云修复;几何修复;模型检索;深度学习
基金项目(Foundation): 国家重点研发计划项目（2016YFC0802107）;; 国家自然科学基金项目（41601409,41501495）;; 北京市自然科学基金项目（8172016）;; 武汉大学测绘遥感信息工程国家重点实验室开放基金资助项目（19E01）;; 北京建筑大学科学研究基金项目（00331616056）;; 无人机倾斜摄像及在教学中的应用研究项目（ZF16095）
作者(Author): 赵江洪,孙铭悦,王殷瑞,窦新铜,张晓光
DOI: 10.16251/j.cnki.1009-2307.2021.01.015
参考文献(References)：
- [1] 王春香,梁亮,王耀,等.三维点云模型孔洞边界识别的研究综述[J].现代制造工程,2019(7):157-162.(WANG Chunxiang,LIANG Liang,WANG Yao,et al.Review of hole boundary recognition in 3D point cloud modell[J].Modern Manufacturing Engineering,2019(7):157-162.)
- [2] YUAN W,KHOT T,HELD D,et al.PCN:point completion network[C]//International Conference on 3D Vision (3DV).Verona,Italy:IEEE,2018:728-737.
- [3] SARKAR K,VARANASI K,STRICKER D.Learning quadrangulated patches for 3D shape parameterization and completion[J].IEEE Computer Society Digital Library,2017(1):383-392.
- [4] LI S,YAO Y,FANG T,et al.Reconstructing thin structures of manifold surfaces by integrating spatial curves[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).[S.l.]:IEEE,2018.
- [5] BOWYER A.Computing dirichlet tessellations[J].The Computer Journal,1978,21(2):168-173.
- [6] WATSON D F.Computing the n-dimensional delaunay tessellation with application to Voronoi polytopes[J].The Computer Journal,1981,24(2):167-172.
- [7] CARR J C,BEATSON R K,CHERRIE J B,et al.Reconstruction and representation of 3D objects with radial basis functions[M].New York:ACM.2001:67-76.
- [8] CHEN C Y,CHENG K Y,LIAO H Y M.A sharpness dependent approach to 3D polygon mesh hole filling[Z].[S.l.:s.n.],2005.
- [9] 晏海平.散乱点云边界提取及孔洞修复算法研究[D].南昌:南昌大学,2014.(YAN Haiping.Research on boundary extraction and hole repair algorithm based on scattered point cloud[D].Nanchang:Nanchang University,2014.)
- [10] DIEBEL J R,THRUN S,BRUNIG M.A bayesian method for probable surface reconstruction and decimation[J].ACM Transactions on Graphics,2006,25(1):39-59.
- [11] PODOLAK J,SHILANE P,GOLOVINSKIY A,et al.A planar-reflective symmetry transform for 3D shapes[J].ACM Transactions on Graphics,2006,25(3):549.
- [12] PAULY M,MITRA N J,WALLNER J,et al.Discovering structural regularity in 3D geometry[J].ACM Transactions on Graphics,2008,27(3):1.
- [13] SIPIRAN I,GREGOR R,SCHRECK T.Approximate symmetry detection in partial 3D meshes[J].Computer Graphics Forum,2014,33(7):1.
- [14] PAULY M,MITRA N J,GIESEN J,GROSS M H.Example-based 3d scan completion[C]//Proceedings of the third Eurographics symposium on Geometry processing.[S.l.:s.n.],2005:23-32.
- [15] LI Y,DAI A,GUIBAS L,et al.Database-assisted object retrieval for real-time 3D reconstruction[J].Computer Graphics Forum,2015,34(2):435-446.
- [16] SUNG M,KIM V G,ANGST R.Data-driven structural priors for shape completion[J].ACM Transactions on Graphics,2015,34(6):1-11.
- [17] GUPTA S,ARBELAEZ P,GIRSHICK R.Aligning 3D models to RGB-D images of cluttered scenes[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Boston,MA,USA:IEEE,2015.
- [18] LI D,SHAO T,WU H,et al.Shape completion from as Single RGBD image[J].IEEE Transactions on Visualization & Computer Graphics,2017,23(7):1809-1822.
- [19] YIN K.Morfit:interactive surface reconstruction from incomplete point clouds with curve-driven topology and geometry control[J].ACM Transactions on Graphics,2014,33(6):1-12.
- [20] SU H,MAJI S,KALOGERAKIS E,et al.Multi-view convolutional neural networks for 3D shape recognition[J].IEEE International Conference on Computer Vision (ICCV).Santiago,Chile:IEEE,2015:2380.
- [21] DANIEL M and SEBASTIAN S.VoxNet:a 3D convolutional neural network for real-time object recognition[EB/OL].[2019-05-09].https://www.researchgate.net.
- [22] QI C R,SU H,KAICHUN M,et al.Point net:deep learning on point sets for 3D classification and segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,HI,USA:IEEE,2017:77-85.
- [23] QI C R,YI L,SU H,et al.PointNet++:deep hierarchical feature learning on point sets in a metric space[EB/OL].[2019-05-09].https://www.researchgate.net/.
- [24] YU L Q,LI X Z,FU C W,et al.PU-Net:point cloud up sampling network[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City,UT,USA:IEEE,2018.
- [25] YANG Y,FENG C,SHEN Y,et al.FoldingNet:point cloud auto-encoder via deep grid deformation[EB/OL].(2017-12-05)[2019-09-6].https://www.researchgate.net/publication/321963241_FoldingNet_Point_Cloud_Auto-encoder_via_Deep_Grid_Deformation.
- [26] HUANG Z,YU Y,XU J,et al.PF-Net:point fractal network for 3D point cloud completion[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR).[S.l.:s.n.],2020.