Please wait a minute...
浙江大学学报(理学版)  2023, Vol. 50 Issue (6): 668-680    DOI: 10.3785/j.issn.1008-9497.2023.06.002
第26届全国计算机辅助设计与图形学学术会议专题     
基于函数映射的二维形状内蕴对称检测算法
刘圣军1,2(),滕子1,王海波1(),刘新儒1
1.中南大学 数学与统计学院,湖南 长沙 410083
2.中南大学 工程建模与科学计算研究所,湖南 长沙 410083
Two-dimensional shape intrinsic symmetry detection algorithm based on functional map
Shengjun LIU1,2(),Zi TENG1,Haibo WANG1(),Xinru LIU1
1.School of Mathematics and Statistics,Central South University,Changsha 410083,China
2.Institute of Engineering Modeling and Scientific Computing,Central South University,Changsha 410083,China
 全文: PDF(4078 KB)   HTML( 4 )
摘要:

针对现有的二维形状内蕴对称检测方法表现欠佳的问题,提出了基于函数映射的二维形状内蕴对称稠密点对应谱优化(FM-2DSISD)方法。首先,设计了对噪声数据鲁棒的稀疏特征对称点对提取算法。其次,利用特征对称点对和函数映射框架,建立了以保持函数映射矩阵每个主子矩阵对角正交性为优化目标的数学模型,证明了该优化目标能保持内蕴对称映射的等距性。借助谱上采样技术,通过频谱域和空间域交替迭代优化函数映射矩阵和逐点映射矩阵。数值实验表明,FM-2DSISD方法对二维光滑形状和噪声形状的检测效果均优于现有检测方法。

关键词: 二维形状内蕴对称谱方法函数映射    
Abstract:

To address the problem that the performance of existing methods is unsatisfactory for detecting intrinsic symmetry in two-dimensional shapes, based on the flexible function mapping framework, we propose a spectral optimization method, named FM-2DSISD, to compute dense point maps for two-dimensional intrinsic symmetric shapes. Firstly, we design an algorithm robust to noise to extract sparse feature symmetry point maps. Secondly, using the feature symmetric point maps and the functional map framework, a mathematical model is developed whose optimization objective is to maintain the diagonality and orthogonality of each principal submatrix of the functional map matrix. We prove that the defined optimization objective can preserve the isometry of intrinsic symmetry maps. To solve the formula, we give an alternating iterative algorithm between the spatial and spectral domains via the spectral up-sampling technique. Numerical experiments show that the proposed algorithm performs better than the state-of-the-art methods on two-dimensional smooth shapes and noisy shapes.

Key words: two-dimensional shape    intrinsic symmetry    spectral method    functional map
收稿日期: 2023-06-12 出版日期: 2023-11-30
CLC:  TP 391.41  
基金资助: 国家自然科学基金资助项目(62172447)
通讯作者: 王海波     E-mail: shjliu.cg@csu.edu.cn;wanghaibo2022@csu.edu.cn
作者简介: 刘圣军(1979—),ORCID:https://orcid.org/0000-0002-3222-5656,男,博士,教授,主要从事几何造型、数字几何处理、数据分析与智能计算研究,E-mail:shjliu.cg@csu.edu.cn.
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
刘圣军
滕子
王海波
刘新儒

引用本文:

刘圣军,滕子,王海波,刘新儒. 基于函数映射的二维形状内蕴对称检测算法[J]. 浙江大学学报(理学版), 2023, 50(6): 668-680.

Shengjun LIU,Zi TENG,Haibo WANG,Xinru LIU. Two-dimensional shape intrinsic symmetry detection algorithm based on functional map. Journal of Zhejiang University (Science Edition), 2023, 50(6): 668-680.

链接本文:

https://www.zjujournals.com/sci/CN/10.3785/j.issn.1008-9497.2023.06.002        https://www.zjujournals.com/sci/CN/Y2023/V50/I6/668

图1  二维形状内蕴对称性检测算法流程
图2  二维图形轮廓特征点提取原理示意
图3  文献[31]中方法在二维形状上得到的错误特征点匹配
图4  关于条件(1)的直观解释
图5  二维形状特征点配对示例
图6  二维内蕴对称检测实验结果
图7  二维光滑轴对称形状不同方法的对称轴可视化结果
图8  二维光滑内蕴对称形状不同方法的可视化检测结果
方法轴对称图形ΔθΔd
SCCCrab1.484 50.9874
Bat0.170 42.476 7
R-DetectorCrab00.484 6
Bat00.349 7
LIP-DetectorCrab00.484 6
Bat00.349 7
FM-2DSISDCrab0.069 10.007 4
Bat0.003 40.043 1
表1  用不同方法得到的不同光滑轴对称图形的对称轴指标
内蕴对称图形FABCICPZoomOutMWPFM-2DSISD
平均值0.072 70.053 30.052 60.055 20.008 4
Beetle0.043 30.043 20.060 20.037 70.011 8
Butterfly0.013 00.009 00.009 50.009 30.004 2
Fly0.157 30.130 50.117 60.174 50.003 9
Human0.070 30.023 20.029 20.030 40.017 5
Octopus0.079 60.060 40.046 30.024 30.004 4
表2  用不同方法得到的不同光滑内蕴对称图形的平均测地误差
图9  二维噪声轴对称形状不同方法的可视化检测结果
图10  二维噪声内蕴对称形状不同方法的可视化检测结果
方法轴对称图形ΔθΔd
SCCCrab0.491 60.330 6
Bat0.846 81.271 0
R-DetectorCrab00.515 6
Bat00.349 7
LIP-DetectorCrab00.515 6
Bat00.349 7
FM-2DSISDCrab0.119 60.089 2
Bat0.033 60.016 8
表3  用不同方法得到的不同噪声轴对称图形的对称轴指标
内蕴对称图形方法
FABCICPZoomOutMWPFM-2DSISD
Beetle(noise)0.043 20.052 50.045 60.040 90.013 6
Butterfly(noise)0.014 50.013 40.013 30.012 40.009 3
Fly(noise)0.106 00.085 10.129 50.103 90.003 7
Human(noise)0.040 70.019 80.061 20.048 20.020 0
Octopus(noise)0.103 20.078 50.047 30.103 80.006 3
平均值0.061 50.049 90.059 40.061 80.010 6
表4  用不同方法得到的不同噪声内蕴对称图形的平均测地误差
三角面片数/个平均测地误差
1 8233.864×10-3
2 5003.737×10-3
3 0003.715×10-3
3 5003.708×10-3
4 0003.684×10-3
表5  用FM-2DSISD方法得到的不同三角面片数图形的平均测地误差
  
(a)1 823个三角面片(b)2 500个三角面片(c)3 000个三角面片(d)3 500个三角面片(e)4 000个三角面片
图11  用Triangle库[35]得到的不同三角面片数图形的三角化可视化结果
方法SCCR-DetectorLIP-DetectorFM-2DSISD

平均运行

时间/s

1.650.070.1034.73
表6  轴对称图形不同方法的平均运行时间
方法FAZoomOutMWPBCICPFM-2DSISD
平均运行时间/s0.030.210.61114.7541.41
表7  内蕴对称图形不同方法的平均运行时间
图12  双手长度相差较大的人体形状及FM-2DSISD方法可视化结果
1 MEHAFFY M, SALINGAROS N. Symmetry in architecture: Toward an overdue reassessment[J]. Symmetry: Culture and Science,2021,32: 311-343. DOI:10.26830/symmetry_2021_3_311
doi: 10.26830/symmetry_2021_3_311
2 GAO L, ZHANG L X, MENG H Y,et al. PRS-Net: Planar reflective symmetry detection net for 3D models[J]. IEEE Transactions on Visualization and Computer Graphics,2020,27(6): 3007-3018. DOI:10.1109/tvcg.2020.3003823
doi: 10.1109/tvcg.2020.3003823
3 LIN H T, LIU Z C, CHEANG C,et al. SAR-Net: Shape alignment and recovery network for category-level 6D object pose and size estimation[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New Orleans: IEEE,2022: 6697-6707. DOI:10.1109/cvpr52688.2022. 00659
doi: 10.1109/cvpr52688.2022. 00659
4 DONATI N, CORMAN E, OVSJANIKOV M. Deep orientation-aware functional maps: Tackling symmetry issues in shape matching[C]// 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New Orleans: IEEE, 2022: 732-741. DOI:10.1109/cvpr52688.2022.00082
doi: 10.1109/cvpr52688.2022.00082
5 NAGAR R, RAMAN S. 3Dsymm: Robust and accurate 3D reflection symmetry detection[J]. Pattern Recognition,2020,107: 107483. DOI:10.1016/j.patcog.2020.107483
doi: 10.1016/j.patcog.2020.107483
6 HUANG J, STOTER J,NAN L. Symmetrization of 2D polygonal shapes using mixed-integer programming[J]. Computer-Aided Design,2023, 163: 103572. DOI:10. 1016/j.cad.2023.103572
doi: 10. 1016/j.cad.2023.103572
7 MITRA N J, GUIBAS L J, PAULY M. Symmetrization[J]. ACM Transactions on Graphics,2007,26(3): 63-70. DOI:10.1145/1276377.1276456
doi: 10.1145/1276377.1276456
8 WU H, CHEN X, LI P,et al. Automatic symmetry detection from brain MRI based on a 2-channel convolutional neural network[J]. IEEE Transactions on Cybernetics,2019,51(9): 4464-4475. DOI:10. 1109/tcyb.2019.2952937
doi: 10. 1109/tcyb.2019.2952937
9 HUANG T, DONG B, LIN J,et al. Symmetry-aware transformer-based mirror detection[C]// Proceedings of the AAAI Conference on Artificial Intelligence. 2023,37(1): 935-943. DOI:10.1609/aaai.v37i1.25173
doi: 10.1609/aaai.v37i1.25173
10 NAGAR R, RAMAN S. Reflection symmetry detection by embedding symmetry in a graph[C]// 2019 IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP). Brighton: IEEE,2019: 2147-2151. DOI:10.1109/icassp.2019.8682412
doi: 10.1109/icassp.2019.8682412
11 GNUTTI A, GUERRINI F, LEONARDI R. Combining appearance and gradient information for image symmetry detection[J]. IEEE Transactions on Image Processing,2021,30: 5708-5723. DOI:10. 1109/tip.2021.3085202
doi: 10. 1109/tip.2021.3085202
12 LEI M, HONG W, ZHAO Z,et al. 3D printing of auxetic metamaterials with digitally reprogrammable shape[J]. ACS Applied Materials & Interfaces,2019,11(25): 22768-22776. DOI:10.1021/acsami. 9b06081
doi: 10.1021/acsami. 9b06081
13 NIU C J, LI J, XU K. Im2struct: Recovering 3D shape structure from a single RBG image[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE,2018: 4521-4529. DOI:10.1109/cvpr.2018.00475
doi: 10.1109/cvpr.2018.00475
14 LEE S, LIU Y. Skewed rotation symmetry group detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,32(9): 1659-1672. DOI:10.1109/tpami.2009.173
doi: 10.1109/tpami.2009.173
15 LEE S, COLLINS R T, LIU Y. Rotation symmetry group detection via frequency analysis of frieze-expansions[C]// 2008 IEEE Conference on Computer Vision and Pattern Recognition. Anchorage: IEEE,2008: 1-8. DOI:10.1109/cvpr. 2008.4587831
doi: 10.1109/cvpr. 2008.4587831
16 NGUYEN T P, NGUYEN T T. Robust detectors of rotationally symmetric shapes based on novel semi-shape signatures[J]. Pattern Recognition,2023,138: 109336. DOI:10.1016/j.patcog.2023.109336
doi: 10.1016/j.patcog.2023.109336
17 AGUILAR W, ALVARADO-GONZALEZ M, GARDUNO E,et al. Detection of rotational symmetry in curves represented by the slope chain code[J]. Pattern Recognition,2020,107: 107421. DOI:10.1016/j.patcog.2020.107421
doi: 10.1016/j.patcog.2020.107421
18 BRIBIESCA E. A geometric structure for two-dimensional shapes and three-dimensional surfaces[J]. Pattern Recognition,1992,25(5): 483-496. DOI:10.1016/0031-3203(92)90047-m
doi: 10.1016/0031-3203(92)90047-m
19 BRIBIESCA E. A measure of tortuosity based on chain coding[J]. Pattern Recognition,2013,46(3): 716-724. DOI:10.1016/j.patcog.2012.09.017
doi: 10.1016/j.patcog.2012.09.017
20 ALVARADO-GONZALEZ M, AGUILAR W, GARDUÑO E,et al. Mirror symmetry detection in curves represented by means of the slope chain code[J]. Pattern Recognition,2019,87: 67-79. DOI:10.1016/j.patcog.2018.10.002
doi: 10.1016/j.patcog.2018.10.002
21 NGUYEN T P, TRUONG H P, NGUYEN T T,et al. Reflection symmetry detection of shapes based on shape signatures[J]. Pattern Recognition,2022,128: 108667. DOI:10.1016/j.patcog.2022.108667
doi: 10.1016/j.patcog.2022.108667
22 NGUYEN T P, NGUYEN X S. Shape measurement using LIP-signature[J]. Computer Vision and Image Understanding,2018,171: 83-94. DOI:10.1016/j.cviu.2018.05.003
doi: 10.1016/j.cviu.2018.05.003
23 TABBONE S, WENDLING L, SALMON J P. A new shape descriptor defined on the Radon transform[J]. Computer Vision and Image Understanding,2006,102(1): 42-51. DOI:10.1016/j.cviu.2005.06.005
doi: 10.1016/j.cviu.2005.06.005
24 OVSJANIKOV M, SUN J, GUIBAS L. Global intrinsic symmetries of shapes[J]. Computer Graphics Forum,2008,27(5): 1341-1348. DOI:10. 1111/j.1467-8659.2008.01273.x
doi: 10. 1111/j.1467-8659.2008.01273.x
25 RUSTAMOV R M. Laplace-Beltrami eigenfunctions for deformation invariant shape representation[C]// Proceedings of the fifth Eurographics Symposium on Geometry Processing. Barcelona:Eurographics Association,2007,257: 225-233.
26 XU K, ZHANG H, TAGLIASACCHI A,et al. Partial intrinsic reflectional symmetry of 3D shapes[J]. ACM Transactions on Graphics,2009: 1-10. DOI:10.1145/1661412.1618484
doi: 10.1145/1661412.1618484
27 XU K, ZHANG H, JIANG W,et al. Multi-scale partial intrinsic symmetry detection[J]. ACM Transactions on Graphics,2012,31(6): 1-11. DOI:10.1145/2366145.2366200
doi: 10.1145/2366145.2366200
28 KIM V G, LIPMAN Y, CHEN X,et al. Möbius transformations for global intrinsic symmetry analysis[J]. Computer Graphics Forum,2010,29(5): 1689-1700. DOI:10.1111/j.1467-8659.2010.01778.x
doi: 10.1111/j.1467-8659.2010.01778.x
29 OVSJANIKOV M, BEN-CHEN M, SOLOMON J,et al. Functional maps: A flexible representation of maps between shapes[J]. ACM Transactions on Graphics,2012,31(4): No.30. DOI:10.1145/2185520.2185526
doi: 10.1145/2185520.2185526
30 WANG H, HUANG H, Group representation of global intrinsic symmetries[J]. Computer Graphics Forum,2017,36(7): 51-61. DOI:10.1111/cgf.13271
doi: 10.1111/cgf.13271
31 NAGAR R, RAMAN S. Fast and accurate intrinsic symmetry detection[C]// 15th European Conference on Computer Vision. Munich: Springer,2018: 433-450. DOI:10. 1007/978-3-030-01246-5_26
doi: 10. 1007/978-3-030-01246-5_26
32 QIAO Y L, GAO L, LIU S Z,et al. Learning-based real-time detection of intrinsic reflectional symmetry[J]. IEEE Transactions on Visualization and Computer Graphics,2023,29(9): 3799-3808. DOI:10.1109/tvcg.2022.3172361
doi: 10.1109/tvcg.2022.3172361
33 REN J, POULENARD A, WONKA P,et al. Continuous and orientation-preserving correspondences via functional maps[J]. ACM Transactions on Graphics,2018,37(6): 1-16. DOI:10.1145/3272127.3275040
doi: 10.1145/3272127.3275040
34 MELZI S, REN J, RODOLA E,et al. ZoomOut: Spectral upsampling for efficient shape correspondence[J]. ACM Transactions on Graphics,2019,38(6): 1-14. DOI:10.1145/3355089.3356524
doi: 10.1145/3355089.3356524
35 SHEWCHUK J R. Triangle: Engineering a 2D quality mesh generator and Delaunay triangulator[C]// Applied Computational Geometry: Towards Geometric Engineering. Berlin: Springer,1996: 203-222. DOI:10. 1007/BFb0014497
doi: 10. 1007/BFb0014497
36 MEYER M, DESBRUN M, SCHRÖDER P,et al. Discrete differential-geometry operators for triangulated 2-manifolds[C]// Visualization and Mathematics III. Berlin: Springer,2003: 35-57. DOI:10.1007/978-3-662-05105-4_2
doi: 10.1007/978-3-662-05105-4_2
37 SARFRAZ M, MASOOD A, ASIM M R. A new approach to corner detection[C]// International Conference on Computer Vision and Graphics. Netherlands: Springer,2006: 528-533. DOI:10.1007/1-4020-4179-9_75
doi: 10.1007/1-4020-4179-9_75
38 SUN J, OVSJANIKOV M, GUIBAS L. A concise and provably informative multi‐scale signature based on heat diffusion[J]. Computer Graphics Forum,2009,28(5): 1383-1392. DOI:10.1111/j.1467-8659. 2009.01515.x
doi: 10.1111/j.1467-8659. 2009.01515.x
39 LATECKI L J, LAKAMPER R, ECKHARDT T. Shape descriptors for non-rigid shapes with a single closed contour[C]// 2000 IEEE Conference on Computer Vision and Pattern Recognition. Hilton Head: IEEE,2000: 424-429. DOI:10.1109/cvpr.2000. 855850
doi: 10.1109/cvpr.2000. 855850
40 ZENG W, GUO R, LUO F,et al. Discrete heat kernel determines discrete Riemannian metric[J]. Graphical Models,2012,74(4): 121-129. DOI:10.1016/j.gmod.2012.03.009
doi: 10.1016/j.gmod.2012.03.009
41 EZUZ D, BEN-CHEN M. Deblurring and denoising of maps between shapes[J]. Computer Graphics Forum, 2017,36(5): 165-174. DOI:10.1111/cgf. 13254
doi: 10.1111/cgf. 13254
42 REN J, MELZI S, WONKA P,et al. Discrete optimization for shape matching[J]. Computer Graphics Forum, 2021,40(5): 81-96. DOI:10.1111/cgf.14359
doi: 10.1111/cgf.14359
43 SEBASTIAN T B, KLEIN P N, KIMIA B B., Recognition of shapes by editing their shock graphs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(5): 550-571. DOI:10.1109/tpami.2004.1273924
doi: 10.1109/tpami.2004.1273924
[1] 叶兴德 江金 生. 双调和方程边值问题的一种新的伪谱逼近[J]. 浙江大学学报(理学版), 1995, 22(1): 13-21.