Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)  2020, Vol. 54 Issue (1): 135-142    DOI: 10.3785/j.issn.1008-973X.2020.01.016
计算机技术、信息工程     
基于等距随机游走图的三维动态曲面对准
程志豪1(),潘翔1,*(),张三元2,任亚楠1
1. 浙江工业大学 计算机科学与技术学院，浙江 杭州 310023
2. 浙江大学 计算机科学与技术学院，浙江 杭州 310058
Three-dimensional dynamic surface alignment based on isometric random walk graph
Zhi-hao CHENG1(),Xiang PAN1,*(),San-yuan ZHANG2,Ya-nan REN1
1. College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou 310023, China
2. College of Computer Science and Technology, Zhejiang University, Hangzhou 310058, China
 全文: PDF(1278 KB)   HTML
摘要：

为了提高三维动态曲面在噪声和遮挡下的对准精度，提出时空等距随机游走图算法. 该算法根据相邻两帧采样点的乘积空间定义图节点，通过时空相邻性进行节点裁剪处理. 以测地距离定义图边约束，将等距映射转化为图稳定性节点选择的随机游走问题. 通过马尔可夫链理论，计算得到最终的对应结果. 通过对不同动态曲面数据库的实验分析表明，该算法针对具有明显噪声和空洞的三维动态曲面能够得到一致性对准关系，性能优于已有算法.
关键词： 三维动态曲面对准;  测地距离;  时空等距随机游走图;  马尔可夫链理论    
Abstract:

A space-time isometric random walk graph was proposed in order to improve the alignment accuracy of three-dimensional dynamic surfaces under noise and occlusion. Graph nodes were defined according to the product space of sampling point sets, and pruning was performed based on spatial-temporal adjacency. The edge weight was defined by the geodesic distance. The isometric mapping problem was formulated into the choice among a random walk graph. The alignment results were computed by Markov chain theory. The experimental results of different dynamic surface databases show that the proposed algorithm can obtain a consistent alignment for three-dimensional dynamic surface with obvious noise and holes. The aligning accuracy of the algorithm is better than the existing algorithms.
Key words: three-dimensional dynamic surface alignment    geodesic distance    space-time isometric random walk graph    Markov chain theory
收稿日期: 2018-11-08 出版日期: 2020-01-05
CLC:  TP 391  
基金资助: 国家重点研发计划重点专项资助项目（2017YFB1002600）；浙江省自然科学基金资助项目（LY19F020031，LQ17F030004）
通讯作者: 潘翔     E-mail: 2512370979@qq.com;panx@zjut.edu.cn
作者简介: 程志豪（1994—），男，硕士生，从事计算机图形学的研究，orcid.org/0000-0002-6695-3803，E-mail： 2512370979@qq.com
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章　　
程志豪
潘翔
张三元
任亚楠

引用本文:

程志豪,潘翔,张三元,任亚楠. 基于等距随机游走图的三维动态曲面对准[J]. 浙江大学学报(工学版), 2020, 54(1): 135-142.

Zhi-hao CHENG,Xiang PAN,San-yuan ZHANG,Ya-nan REN. Three-dimensional dynamic surface alignment based on isometric random walk graph. Journal of ZheJiang University (Engineering Science), 2020, 54(1): 135-142.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2020.01.016    或    http://www.zjujournals.com/eng/CN/Y2020/V54/I1/135

图 1  时空等距游走映射算法流程图
图 2  采样频率对本文算法精度的影响
图 3  采样频率对本文算法运行效率的影响
图 4  部分三维模型的表面采样结果
图 5  DFAUST数据库各算法定量分析
图 6  DFAUST数据库各算法可视化匹配结果（从上至下依次为IRWG算法、KM算法、RAVAC算法、LRST算法、GMDSA算法、CCM算法）
图 7  SCPA数据库各算法的定量分析
图 8  SCPA数据库各算法可视化匹配结果（从上至下依次为IRWG算法、KM算法、RAVAC算法、LRST算法、GMDSA算法、CCM算法）
图 9  LRST算法与IRWG算法结果比较
图 10  MVPS数据库各算法可视化匹配结果（从上至下依次为IRWG算法，KM算法，RAVAC算法，LRST算法，CCM算法）
算法 Jay Saskia Abhijeet
KM 0.294 1 0.288 5 0.259 5
IRWG 0.062 1 0.085 7 0.077 9
RAVAC 0.331 2 0.294 4 0.286 9
LRST 0.226 7 0.199 8 0.235 1
GMDSA N/A N/A N/A
CCM 0.385 6 0.351 3 0.376 9
表 1  MVPS数据库各算法定量分析
帧数 每帧顶点数 KM IRWG RAVAC LRST GMDSA CCM
290 6 890 38 461 1 131 3 051 1 614 N/A 859
500 3 463 34 716 490 3 379 1 361 N/A 417
145 4 970 17 332 754 2 185 1 054 N/A 168
表 2  不同对应算法的运行时间
1 GUO K, XU F, WANG Y, et al. Robust non-rigid motion tracking and surface reconstruction using l0 regularization [C] // IEEE International Conference on Computer Vision. Santiago: IEEE, 2015: 3083-3091.
2 SAHILLIO?LU Y, YEMEZ Y Partial 3D correspondence from shape extremities[J]. Computer Graphics Forum, 2015, 33 (6): 63- 76
3 BRONSTEIN A M, BRONSTEIN M M, KIMMEL R Generalized multidimensional scaling: a framework for isometry-invariant partial surface matching[J]. Proceedings of the National Academy of Sciences of the United States of America, 2006, 103 (5): 1168- 72
doi: 10.1073/pnas.0508601103
4 潘翔, 王学成, 张三元 基于等距二分图的三维模型局部对齐[J]. 计算机辅助设计与图形学学报, 2016, (3): 480- 487
PAN Xiang, WANG Xue-cheng, ZHANG San-yuan 3D partial correspondence based on isometric bipartite graph[J]. Journal of Computer-Aided Design and Computer Graphics, 2016, (3): 480- 487
doi: 10.3969/j.issn.1003-9775.2016.03.014
5 郭梦丽, 达飞鹏, 邓星, 等 基于关键点和局部特征的三维人脸识别[J]. 浙江大学学报: 工学版, 2017, 51 (03): 584- 589
GUO Meng-li, DA Fei-peng, DENG Xing, et al 3D face recognition based on keypoints and local feature[J]. Journal of Zhejiang University: Engineering Science, 2017, 51 (03): 584- 589
6 OVSJANIKOV M, QUENTIN M, FACUNDO M, et al One point isometric matching with the heat kernel[J]. Computer Graphics Forum, 2010, 29 (5): 1555- 1564
doi: 10.1111/j.1467-8659.2010.01764.x
7 COSMO L, ALBARELLI A, CREMERS D Consistent partial matching of shape collections via sparse modeling[J]. Computer Graphics Forum, 2017, 36 (1): 209- 221
doi: 10.1111/cgf.12796
8 RODOL? E, COSMO L, BRONSTEIN M M, et al Partial functional correspondence[J]. Computer Graphics Forum, 2017, 36 (1): 222- 236
doi: 10.1111/cgf.12797
9 L?HNER Z, VESTNER M, BOYARSKI A, et al. Effi-cient deformable shape correspondence via kernel matching [C] // International Conference on 3d Vision. Qingdao: [s.n.], 2017: 517-526.
10 LITANY O, RODOL? E, BRONSTEIN A M, et al Fully spectral partial shape matching[J]. Computer Graphics Forum, 2017, 36 (2): 247- 258
doi: 10.1111/cgf.13123
11 AFLALO Y, DUBROVINA A, KIMMEL R Spectral generalized multi-dimensional scaling[J]. International Journal of Computer Vision, 2013, 118 (3): 1- 13
12 战江涛, 刘强, 柴春雷 基于三维模型与Gabor小波的人脸特征点跟踪方法[J]. 浙江大学学报: 工学版, 2011, 45 (1): 30- 36
ZHAN Jiang-tao, LIU Qiang, CHAI Chun-lei Facial feature tracking using three-dimensional model and Gabor wavelet[J]. Journal of Zhejiang University: Engineering Science, 2011, 45 (1): 30- 36
13 TUNG T, MATSUYAMA T Geodesic mapping for dynamic surface alignment[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36 (5): 901- 913
doi: 10.1109/TPAMI.2013.179
14 MUSTAFA A, KIM H, HILTON A. 4D match trees for non-rigid surface alignment [C] // European Conference on Computer Vision. Amsterdan: Springer, 2016: 213-229.
15 LIAO B, XIAO C, PANG Z Efficient feature tracking of time-varying surfaces using multi-scale motion flow propagation[J]. Computer-Aided Design, 2013, 45: 1394- 1407
doi: 10.1016/j.cad.2013.06.015
16 SAHILLIO?LU Y, YEMEZ Y Scale normalization for isometric shape matching[J]. Computer Graphics Forum, 2012, 31 (7): 2233- 2240
doi: 10.1111/j.1467-8659.2012.03216.x
17 SENETA E. Non-negative matrices and Markov chains [M]. New York: Springer, 2006.
18 CHO M, LEE J, LEE K M. Reweighted random walks for graph matching [C] // European Conference on Computer Vision. Crete: Springer, 2010: 492-505.
19 BOGO F, ROMERO J, PONSMOLL G, et al. Dynamic FAUST: registering human bodies in motion [C] // IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 5573-5582.
20 STARCK J, HILTON A Surface capture for performance-based animation[J]. IEEE Computer Graphics and Applications, 2007, 27 (3): 21- 31
doi: 10.1109/MCG.2007.68
21 VLASIC D, PEERS P, BARAN I, et al Dynamic shape capture using multi-view photometric stereo[J]. ACM Transactions on Graphics (TOG), 2009, 28 (5): 1- 11
[1] 郑守国,张勇德,谢文添,樊虎,王青. 基于数字孪生的飞机总装生产线建模[J]. 浙江大学学报(工学版), 2021, 55(5): 843-854.
[2] 张师林,马思明,顾子谦. 基于大边距度量学习的车辆再识别方法[J]. 浙江大学学报(工学版), 2021, 55(5): 948-956.
[3] 宋鹏,杨德东,李畅,郭畅. 整体特征通道识别的自适应孪生网络跟踪算法[J]. 浙江大学学报(工学版), 2021, 55(5): 966-975.
[4] 蔡君,赵罡,于勇,鲍强伟,戴晟. 基于点云和设计模型的仿真模型快速重构方法[J]. 浙江大学学报(工学版), 2021, 55(5): 905-916.
[5] 王虹力,郭斌,刘思聪,刘佳琪,仵允港,於志文. 边端融合的终端情境自适应深度感知模型[J]. 浙江大学学报(工学版), 2021, 55(4): 626-638.
[6] 张腾,蒋鑫龙,陈益强,陈前,米涛免,陈彪. 基于腕部姿态的帕金森病用药后开-关期检测[J]. 浙江大学学报(工学版), 2021, 55(4): 639-647.
[7] 郑英杰,吴松荣,韦若禹,涂振威,廖进,刘东. 基于目标图像FCM算法的地铁定位点匹配及误报排除方法[J]. 浙江大学学报(工学版), 2021, 55(3): 586-593.
[8] 雍子叶,郭继昌,李重仪. 融入注意力机制的弱监督水下图像增强算法[J]. 浙江大学学报(工学版), 2021, 55(3): 555-562.
[9] 于勇,薛静远,戴晟,鲍强伟,赵罡. 机加零件质量预测与工艺参数优化方法[J]. 浙江大学学报(工学版), 2021, 55(3): 441-447.
[10] 胡惠雅,盖绍彦,达飞鹏. 基于生成对抗网络的偏转人脸转正[J]. 浙江大学学报(工学版), 2021, 55(1): 116-123.
[11] 陈杨波,伊国栋,张树有. 基于点云特征对比的曲面翘曲变形检测方法[J]. 浙江大学学报(工学版), 2021, 55(1): 81-88.
[12] 段有康,陈小刚,桂剑,马斌,李顺芬,宋志棠. 基于相位划分的下肢连续运动预测[J]. 浙江大学学报(工学版), 2021, 55(1): 89-95.
[13] 张太恒,梅标,乔磊,杨浩杰,朱伟东. 纹理边界引导的复合材料圆孔检测方法[J]. 浙江大学学报(工学版), 2020, 54(12): 2294-2300.
[14] 梁栋,刘昕宇,潘家兴,孙涵,周文俊,金子俊一. 动态背景下基于自更新像素共现的前景分割[J]. 浙江大学学报(工学版), 2020, 54(12): 2405-2413.
[15] 晋耀,张为. 采用Anchor-Free网络结构的实时火灾检测算法[J]. 浙江大学学报(工学版), 2020, 54(12): 2430-2436.