›› 2013,Vol. 28 ›› Issue (3): 479-489.doi: 10.1007/s11390-013-1349-x

所属专题： Artificial Intelligence and Pattern Recognition； Computer Graphics and Multimedia

• Special Section on Selected Paper from NPC 2011 • 上一篇    下一篇

由流形约束实现人脸知识迁移的三维卡通重建方法

Jun-Fa Liu1,2 (刘军发), Senior Member, CCF, Member, ACM, IEEE, Wen-Jing He1,2,3 (何文静), Tao Chen4 (陈涛), and Yi-Qiang Chen1,2 (陈益强), Senior Member, CCF, Member, ACM, IEEE   

  1. 1. Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China;
    2. Beijing Key Laboratory of Mobile Computing and Pervasive Device, Beijing 100190, China;
    3. University of Chinese Academy of Sciences, Beijing 100049, China;
    4. School of Electronic Information and Automation, Chongqing University of Technology, Chongqing 400054, China
  • 收稿日期:2012-02-07 修回日期:2013-01-23 出版日期:2013-05-05 发布日期:2013-05-05
  • 作者简介:Jun-Fa Liu received the Ph.D. degree in computer application technology from the Institute of Computing Technology (ICT), Chinese Academy of Sciences (CAS), Beijing, in 2009. From July 2009, he has been working in ICT, CAS, and now is an associate professor. His current research interests include machine learning, computational intelligence, human computer interaction and pervasive computing.
  • 基金资助:

    This work was supported by the National Natural Foundation of China under Grant Nos. 61070110 and 61173066.

Manifold Constrained Transfer of Facial Geometric Knowledge for 3D Caricature Reconstruction

Jun-Fa Liu1,2 (刘军发), Senior Member, CCF, Member, ACM, IEEE, Wen-Jing He1,2,3 (何文静), Tao Chen4 (陈涛), and Yi-Qiang Chen1,2 (陈益强), Senior Member, CCF, Member, ACM, IEEE   

  1. 1. Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China;
    2. Beijing Key Laboratory of Mobile Computing and Pervasive Device, Beijing 100190, China;
    3. University of Chinese Academy of Sciences, Beijing 100049, China;
    4. School of Electronic Information and Automation, Chongqing University of Technology, Chongqing 400054, China
  • Received:2012-02-07 Revised:2013-01-23 Online:2013-05-05 Published:2013-05-05
  • Contact: 10.1007/s11390-013-1349-x
  • Supported by:

    This work was supported by the National Natural Foundation of China under Grant Nos. 61070110 and 61173066.

PDF (PC)

1798

三维卡通人脸是一种重要的资源,可以广泛应用于在线游戏、网络虚拟社区、卡通动画影视等领域。目前三维模型的制作可以基于专用软件,如MAYA,3DS MAX等,但这是一项非常繁琐的手工劳动,同时,三维卡通人脸创作要求制作者具有足够的艺术背景,以确保创作的质量效果。但漫画艺术家们在进行创作时,主要以二维的卡通为主,现实中存在的三维卡通人脸资源十分有限。目前已有一些研究工作对相关问题进行了研究以实现自动化生成三维卡通人脸,如采用PCA方法等。但是这些工作在选择统计模型时采用线性模型,对本文复杂的三维卡通人脸数据尤其是人脸侧面细节方面进行模拟时,尚存在明显的不足。本文提出一种自动化生成方法,其核心思想为：既然二维卡通人脸是三维卡通人脸在二维平面的投影,可以认为二者是同一种特征的不同“视图(View)”,同时又由于二者均具有显著的流形特征,则可以认为二者来源于同一潜在流形分布。基于这一思想,本文提出一种三维重建方法,基于一批真实的二维卡通人脸,重建相应的一批三维卡通人脸。首先,根据100个在MAYA环境下手工制作的三维卡通初始化全体三维数据集。其次,对二维卡通与三维卡通进行约束降维,使二者在某一低维流形嵌入上具有相同流形特征,即二者共享此流形。基于此流形进行流形升维计算,对三维卡通数据集进行重建,因为该流形已包含二维数据信息,因此对三维卡通数据集的重建是对数据集的进一步修正和知识迁移。然后再次对二维数据集与三维数据集进行约束流形降维计算和流形升维计算,进入迭代步骤,直至算法收敛。对比试验表明,该方法能够有效地重建三维卡通人脸,与主成分分析方法相比更能体现出原始二维卡通人脸的风格。

Abstract: 3D caricatures are important attractive elements of the interface in virtual environment such as online game. However, very limited 3D caricatures exist in the real world. Meanwhile, creating 3D caricatures manually is rather costly, and even professional skills are needed. This paper proposes a novel and effective manifold transfer algorithm to reconstruct 3D caricatures according to their original 2D caricatures. We first manually create a small dataset with only 100 3D caricature models and use them to initialize the whole 3D dataset. After that, manifold transfer algorithm is carried out to refine the dataset. The algorithm comprises of two steps. The first is to perform manifold alignment between 2D and 3D caricatures to get a "standard" manifold map; the second is to reconstruct all the 3D caricatures based on the manifold map. The proposed approach utilizes and transfers knowledge of 2D caricatures to the target 3D caricatures well. Comparative experiments show that the approach reconstructs 3D caricatures more effectively and the results conform more to the styles of the original 2D caricatures than the Principal Components Analysis (PCA) based method.

[1] Sadimon S B, Sunar M S, Mohamad D, Haron H. Computer generated caricature: A survey. In Proc. the 2010 Int. Conf. Cyberworlds Pages, Oct. 2010, pp.383-390.

[2] Kondo T, Murakami K, Koshimizu H. From coarse to fine correspondence of 3-D facial images and its application to 3-D facial caricaturing. In Proc. Int. Conf. Recent Advances in 3-D Digital Imaging and Modeling, May 1997, pp.283-288.

[3] Fujiwara T, Koshimizu H, Fujimura K et al. A method for 3D face modeling and caricatured figure generation. In Proc. ICME 2002, Aug. 2002, pp.137-140.

[4] Shadbolt A. From 2D photographs to 3D caricatures. Technical Report, Department of Computer Science, University of Sheffield, 2003.

[5] Lim Y K, Fedorov A, Kim S D. 3D caricature generation system on the mobile handset using a single photograph. In Proc. ICPPW2007, Sept. 2007, Article No.37.

[6] Clarke L, Chen M, Mora B. Automatic generation of 3D caricatures based on artistic deformation styles. IEEE Trans. Visualization and Computer Graphics, 2011, 17(6): 808-821.

[7] Liu J F, Chen Y Q, Miao C Y et al. Semi-supervised learning in reconstructed manifold space for 3D caricature generation. Computer Graphics Forum, 2009, 28(8): 2104-2116.

[8] Li P F, Chen Y Q, Liu J F et al. 3D caricature generation by nonlinear manifold learning. In Proc. ICME, June 2008, pp.941-944.

[9] Fu G H, Chen Y Q, Liu J F et al. Interactive expressive 3D caricatures design. In Proc. ICME, June 2008, pp.965-968.

[10] Akleman E, Reisch J. Modeling expressive 3D caricatures. In Proc. the 31st SIGGRAPH, Aug. 2004, p.61.

[11] Akimoto T, Suenaga Y, Wallace R S. Automatic creation of 3D facial models. Computer Graphics and Applications, 1993, 13(5): 16-22.

[12] Pighin F, Hecker J, Lischinski D, Szeliski R, Shalesin D H. Synthesizing realistic facial expressions from photographs. In Proc. the 25th SIGGRAPH, July 1998, pp.75-84.

[13] Liu Z, Zhang Z, Jacobs C, Cohen M. Rapid modeling of animated faces from video. Technical Report, MSR-TR-2000-11, Microsoft Research, February 2000.

[14] Blanz V, Vetter T. A morphable model for the synthesis of 3D-faces. In Proc. the 26th SIGGRAPH, Aug. 1999, pp.187194.

[15] Jiang D, Hu Y, Yan S et al. Efficient 3D reconstruction for face recognition. Pattern Recognition, 2005, 38(6): 787-798.

[16] Lee J, Moghaddam B, Pfister H, Machiraju R. Silhouettebased 3D face shape recovery. In Proc. Graphics Interface 2003, June 2003, pp.21-30.

[17] Nandy D, Ben-Arie J. Shape from recognition: A novel approach for 3-D face shape recovery. IEEE Transactions on Image Processing, 2001, 10(2): 206-217.

[18] Andreetto M, Brusco N, Cortelazzo G M. Automatic 3D modeling of textured cultural heritage objects. IEEE Transactions on Image Processing, 2004, 13(3): 354-369.

[19] Roy-Chowdhury A K, Chellappa R. Statistical bias in 3-D reconstruction from a monocular video. IEEE Transactions on Image Processing, 2005, 14(8): 1057-1062.

[20] Ham J, Lee D, Saul L. Semisupervised alignment of manifolds. In Proc. the 10th International Workshop on Artificial Intelligence and Statistics, Jan. 2005, pp.187-194.

[21] Roweis S T, Saul L K. Nonlinear dimensionality reduction by locally linear embedding. Science, 2000, 290(5500): 23232326.

[22] Oliver M A, Webster R. Kriging: A method of interpolation for geographical information system. INT. J. Geographical Information Systems, 1990, 4(3): 313-332.

[23] Dempster A P, Laird N M, Rubin D B. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society, 1977, 39(1): 1-38.

[24] Belkin M, Niyogi P. Laplacian eigenmaps for dimensionality reduction and data representation. Neural Computation, 2003, 15(6): 1373-1396.
No related articles found!
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] 冯玉琳;. Hierarchical Protocol Analysis by Temporal Logic[J]. , 1988, 3(1): 56 -69 .
[2] 张钹; 张铃;. A Relation Matrix Approach to Labelling Temporal Relations in Scheduling[J]. , 1991, 6(4): 339 -346 .
[3] . 判定XPath片段包含关系的隐式条件同态映射[J]. , 2007, 22(5): 736 -748 .
[4] . 基于单幅深度图像的树枝重建研究[J]. , 2007, 22(6): 846 -858 .
[5] Feng Xu Jing Pan Wen Lu. 一个开放环境中基于信任的软件系统可信度评估方法[J]. , 2009, 24(2): 373 -385 .
[6] Jing Zhou, Shan-Feng Zhu, Xiaodi Huang, Yanchun Zhang. 基于半监督学习整合多种时间序列距离度量方法的时间序列聚类[J]. , 2015, 30(4): 859 -873 .
[7] Tao Xie. Preface[J]. , 2016, 31(5): 849 -850 .
[8] Dongchul Park, Weiping He, David H. C. Du. 多布隆过滤器热数据识别:区块级别决策vs I/O请求级别决策[J]. , 2018, 33(1): 79 -97 .
版权所有 © 《计算机科学技术学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持：support@magtech.com.cn
总访问量：