基于逆运动学和重构式ICA的人体运动风格分析与合成

蓝荣祎; 孙怀江

doi:10.3724/SP.J.1004.2014.01135

基于逆运动学和重构式ICA的人体运动风格分析与合成

doi: 10.3724/SP.J.1004.2014.01135

1.
南京理工大学计算机科学与工程学院南京 210094

基金项目:

高等学校学科创新引智计划（B13022），南京理工大学自主科研专项计划资助项目（2011YBXM79）资助

详细信息

作者简介:
蓝荣祎南京理工大学博士研究生.2008 年获得南京理工大学学士学位. 主要研究方向为模式识别，人体运动捕获数据重用.E-mail：brianlanbo@gmail.com

计量
- 文章访问数: 1795
- HTML全文浏览量: 78
- PDF下载量: 1131
- 被引次数: 0
出版历程
- 收稿日期: 2013-02-05
- 修回日期: 2013-07-04
- 刊出日期: 2014-06-20

Style Analysis and Human Locomotion Synthesis Based on Inverse Kinematics and Reconstructive ICA

1.
School of Computer Science and Engineering, Nanjing University of Science & Technology, Nanjing 210094

Funds:

Supported by Innovation and Talent Introduction Funding Program of Colleges & Universities (B13022), Independent Research Funding Program of Nanjing University of Science and Technology (2011YBXM79)

摘要

摘要: 使用独立成分分析（Independent component analysis，ICA）来建模运动风格、合成风格化的人体运动，是一种有效且有前景的手段.为了避免现有方法在设定独立成分个数或子空间结构时的人为影响，并提高风格成分的质量，提出一种基于重构式独立成分分析的运动风格分析方法.由于放弃了混合矩阵的正交性约束，一方面，拥有了更多的自由度来表示各独立成分；另一方面，利用特征的过完备性以及自身在特征选择时的稀疏特性，能够自动地确立独立成分数目.此外，通过结合基于主测地线分析的逆运动学与运动过渡技术，该方法能够合成包含多种风格、任意长度的行走运动，同时还能通过编辑特定帧的人体姿势来约束合成的结果.实验结果表明，该方法能够有效地分析出行走、跳跃和踢腿等运动中代表风格的独立成分，并根据用户对风格的编辑，实时地生成自然、平滑的运动.
- 独立成分分析 /
- 运动捕获数据 /
- 人体运动合成 /
- 逆运动学 /
- 风格分析
Abstract: Independent component analysis (ICA) has been proven to be an effective and promising way to model motion styles and synthesize stylistic human motions. To avoid heuristic attempting on the number of independent components or the structure of independent subspaces, we propose a reconstructive ICA based method to analyze motion styles. By alleviating the orthogonal constraint to the mixing matrix, overcomplete features can be obtained, which enables more degrees of freedom to represent each independent component, and also, the sparse effect during feature selection automatically determines the number of independent components. Furthermore, by introducing the principal geodesic analysis and motion transition technique, the proposed method is able to generate motion sequences containing multiple styles and of arbitrary lengths, and refinement according to specific pose constraints can also be made. The experimental results demonstrate that the proposed method can effectively extract the style component of locomotion, jumping, kicking, etc., and produce natural, smooth motion that responds to the editing of styles in real-time.
- Independent component analysis (ICA) /
- motion capture data /
- human motion synthesis /
- inverse kinematics /
- style analysis

HTML全文

参考文献(33)

[1]	Moeslund B T, Hilton A, Krüger Volker. A survey of advances in vision-based human motion capture and analysis. Computer Vision and Image Understanding, 2006, 104(2-3): 90-126
[2]	Zhang Li-Ge, Bi Shu-Sheng, Gao Jin-Lei. Human motion data acquiring and analyzing method for humanoid robot motion designing. Acta Automatica Sinica, 2010, 36(1): 107-112(张利格, 毕树生, 高金磊. 仿人机器人复杂动作设计中人体运动数据提取及分析方法. 自动化学报, 2010, 36(1): 107-112)
[3]	Wang J M, Fleet D J, Hertzmann A. Multifactor Gaussian process models for style-content separation. In: Proceedings of the 24th International Conference on Machine Learning. New York, USA: ACM Press, 2007. 975-982
[4]	Min J Y, Liu H J, Chai J X. Synthesis and editing of personalized stylistic human motion. In: Proceedings of the 2010 ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games. New York, USA: ACM Press, 2010. 39-46
[5]	Liu G D, Xu M L, Pan Z G, Rhalibi A E. Human motion generation with multifactor models. Computer Animation and Virtual Worlds, 2011, 22(4): 351-359
[6]	Ikemoto L, Arikan O, Forsyth D. Generalizing motion edits with Gaussian processes. ACM Transactions on Graphics (TOG), 2009, 28(1): 1-12
[7]	Hsu E, Pulli K, Popović J. Style translation for human motion. ACM Transactions on Graphics (TOG), 2005, 24(3): 1082-1089
[8]	Wu Xiao-Mao, Ma Li-Zhuang, Zheng Can, Chen Yan-Yun. Algorithm for real-time style transfer for human motion. Journal of System Simulation, 2006, 18(S1): 393-395, 399(吴小亻!!!!毛, 马利庄, 郑灿, 陈彦云. 实时人体运动风格传输算法. 系统仿真学报, 2006, 18(S1): 393-395, 399)
[9]	Rose C, Cohen M F, Bodenheimer B. Verbs and adverbs: multidimensional motion interpolation. IEEE Computer Graphics and Applications, 1998, 18(5): 32-40
[10]	Mukai T, Kuriyama S. Geostatistical motion interpolation. ACM Transactions on Graphics (TOG), 2005, 24(3): 1062-1070
[11]	Ma W L, Xia S H, Hodgins J K, Yang X, Li C P, Wang Z Q. Modeling style and variation in human Motion. In: Proceedings of the 2010 ACM SIGGRAPH/Eurographics Symposium on Computer Animation. Aire-la-Ville, Switzerland: Eurographics Association, 2010. 21-30
[12]	Shapiro A, Cao Y, Faloutsos P. Style components. In: Proceedings of the 2006 Graphics Interface. Toronto, Canada: Canadian Information Processing Society, 2006. 33-39
[13]	Liu G D, Pan Z G, Lin Z Y. Style subspaces for character animation. Computer Animation and Virtual Worlds, 2008, 19(3-4): 199-209
[14]	Liu Geng-Dai, Xu Ming-Liang, Zhang Ming-Min. Human motion synthesis based on independent spatio-temporal feature space. Chinese Journal of Computers, 2011, 34(3): 464-472(刘更代, 徐明亮, 张明敏. 基于独立时空特征空间的人体运动合成. 计算机学报, 2011, 34(3): 464-472)
[15]	Kim Y, Neff M. Component-based locomotion composition. In: Proceedings of the 2012 ACM SIGGRAPH/Eurographics Symposium on Computer Animation. Aire-la-Ville, Switzerland: Eurographics Association, 2012. 165-173
[16]	Le Q V, Karpenko A, Ngiam J, Ng A Y. ICA with reconstruction cost for efficient overcomplete feature learning. In: Proceedings of 25th Annual Conference on Neural Information Processing Systems (NIPS). Granada, Spain: NIPS, 2011. 1017-1025
[17]	Liu C K, Hertzmann A, Popović Z. Learning physics-based motion style with nonlinear inverse optimization. ACM Transactions on Graphics (TOG), 2005, 24(3): 1071-1081
[18]	De Lasa M, Mordatch I, Hertzmann A. Feature-based locomotion controllers. ACM Transactions on Graphics (TOG), 2010, 29(4): 131
[19]	Grochow K, Martin S L, Hertzmann A, Popović Z. Style-based inverse kinematics. ACM Transactions on Graphics (TOG), 2004, 23(3): 522-531
[20]	Vasilescu M A O. Human motion signatures: analysis, synthesis, recognition. In: Proceedings of the 16th International Conference on Pattern Recognition. Quebec City, Quebec, Canada: IEEE, 2002, 3: 456-460
[21]	Hertzmann A, Jacobs C E, Oliver N, Curless B, Salesin D H. Image analogies. In: Proceedings of the 28th Annual Conference on Computer Graphics and Interactive Techniques. New York, USA: ACM Press, 2001. 327-340
[22]	Freeman W T, Tenenbaum J B, Pasztor E C. Learning style translation for the lines of a drawing. ACM Transactions on Graphics (TOG), 2003, 22(1): 33-46
[23]	Liu Chang-Song, Ding Xiao-Qing. Study of character recognition using writing style consistent. Acta Automatica Sinica, 2007, 33(11): 1121-1127(刘长松, 丁晓青. 利用字形风格约束的字符识别研究. 自动化学报, 2007, 33(11): 1121-1127)
[24]	Hyvärinen A, Oja E. Independent component analysis: algorithms and applications. Neural Networks, 2000, 13(4-5): 411-430
[25]	Tournier M, Wu X, Courty N, Arnaud E, Revéret L. Motion compression using principal geodesics analysis. Computer Graphics Forum, 2009, 28(2): 355-364
[26]	Liu Geng-Dai. Study on Synthesis of Character Animation and Its Stylization [Ph.D. dissertation], Zhejiang University, China, 2009(刘更代. 人体动画合成及其风格化处理研究 [博士学位论文], 浙江大学, 中国, 2009)
[27]	Liu Geng-Dai, Pan Zhi-Geng, Cheng Xi, Xu Ming-Liang. Stylistic human motion synthesis with low-dimensional motion model and inverse kinematics. Journal of Computer-Aided Design and Computer Graphics, 2010, 22(1): 145-151(刘更代, 潘志庚, 程熙, 徐明亮. 结合低维运动模型和逆运动学的风格化人体运动合成. 计算机辅助设计与图形学学报, 2010, 22(1): 145-151)
[28]	Fletcher P T, Lu C, Joshi S. Statistics of shape via principal geodesic analysis on Lie groups. In: Proceedings of 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. NC, USA: IEEE, 2003. I-95-I-101
[29]	Said S, Courty N, Le Bihan N. Sangwine S J. Exact principal geodesic analysis for data on SO(3). In: Proceedings of the 15th European Signal Processing Conference. Pozna\'{n, Poland: EURASIP, 2007. 1700-1705
[30]	Johnson M P. Exploiting Quaternions to Support Expressive Interactive Character Motion [Ph.D. dissertation], Massachusetts Institute of Technology, USA, 2003
[31]	Le Q V, Ranzato M A, Monga R, Devin M, Chen K, Corrado G S, Dean J, Ng A Y. Building high-level features using large scale unsupervised learning. In: Proceedings of the 29th International Conference on Machine Learning. Ediburgh, Scotland, 2012. 81-88
[32]	Moré；J J. The Levenberg-Marquardt algorithm: implementation and theory. Numerical Analysis. Berlin Heidelberg: Springer, 1978. 105-116
[33]	Gleicher M. Retargetting motion to new characters. In: Proceedings of the 25th Annual Conference on Computer Graphics and Interactive Techniques. New York, USA: ACM Press, 1998. 33-42

施引文献

资源附件(0)

访问统计

点击查看大图

计量

文章访问数: 1795
HTML全文浏览量: 78
PDF下载量: 1131
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

留言板

基于逆运动学和重构式ICA的人体运动风格分析与合成

doi: 10.3724/SP.J.1004.2014.01135

作者简介:
蓝荣祎南京理工大学博士研究生.2008 年获得南京理工大学学士学位. 主要研究方向为模式识别，人体运动捕获数据重用.E-mail：brianlanbo@gmail.com

计量

Style Analysis and Human Locomotion Synthesis Based on Inverse Kinematics and Reconstructive ICA

计量

目录

留言板

基于逆运动学和重构式ICA的人体运动风格分析与合成

doi: 10.3724/SP.J.1004.2014.01135

作者简介: 蓝荣祎 南京理工大学博士研究生.2008 年获得南京理工大学学士学位. 主要研究方向为模式识别，人体运动捕获数据重用.E-mail：brianlanbo@gmail.com

计量

出版历程

Style Analysis and Human Locomotion Synthesis Based on Inverse Kinematics and Reconstructive ICA

计量

出版历程

目录

作者简介:
蓝荣祎南京理工大学博士研究生.2008 年获得南京理工大学学士学位. 主要研究方向为模式识别，人体运动捕获数据重用.E-mail：brianlanbo@gmail.com