张量描述下的多姿态多表情人脸合成方法

doi:10.3724/SP.J.1087.2012.00256

计算机应用 ›› 2012, Vol. 32 ›› Issue (01): 256-260.DOI: 10.3724/SP.J.1087.2012.00256

张量描述下的多姿态多表情人脸合成方法

吕煊¹,²,王志成¹,²,赵卫东¹,²

1. 同济大学计算机辅助设计(CAD)研究中心，上海 200092
2. 同济大学企业数字化技术教育部工程研究中心，上海 200092

收稿日期:2011-06-27 修回日期:2011-08-15 发布日期:2012-02-06 出版日期:2012-01-01
通讯作者: 王志成
作者简介:吕煊(1982-)，男，山东淄博人,博士研究生,主要研究方向：数字图像处理、计算机视觉；王志成(1975-)，男，江苏泰兴人，副研究员，主要研究方向：数字图像处理；赵卫东(1965-)，男，山东青岛人，研究员，主要研究方向：制造业信息化。

Multi-pose and expression face synthesis method based on tensor representation

Lǚ Xuan¹,²,WANG Zhi-cheng¹,²,ZHAO Wei-dong¹,²

1. CAD Research Center, Tongji University, Shanghai 200092, China
2. Engineering Research Center for Enterprise Digital Technology of Ministry of Education, Tongji University, Shanghai 200092, China

Received:2011-06-27 Revised:2011-08-15 Online:2012-02-06 Published:2012-01-01
Contact: WANG Zhi-cheng

摘要/Abstract

摘要： 为了从一幅人脸图像中合成出该人脸其他姿态和表情下的图像，提出了一种基于张量子空间的多姿态人脸表情合成方法。首先，用标记过特征点的人脸图像集构造四维纹理特征张量和形状张量；其次，通过张量分解得到核张量以及各维的投影子空间(人物标识、表情、姿态、特征维)；最后应用核张量以及表情、姿态子空间构造新的张量用于姿态、表情的合成，在合成新人脸图像的时候充分利用了影响人脸的各因素间的内在关系。实验结果表明，所提方法可以利用一张已知表情和姿态的人脸图合成出自然合理的其他姿态表情下的该人脸图像。

关键词: 张量分解, 表情、姿态合成, 模式子空间

Abstract: To synthesize facial pose and expression images simultaneously from one image, a tensor-based subspace projection method for synthesizing multi-pose and expression face images was proposed. Firstly, the forth order texture tensor and shape tensor were created from the feature annotated images respectively. Then a tucker tensor decomposition technique was applied to build projection subspaces (person, expression, pose and feature subspaces). Core tensors, expressions, poses and feature subspaces were organized into a new tensor properly which was used for synthesizing new facial poses and expressions. The proposed method took full advantage of the intrinsic relationship among the facial affected various factors. The experimental results show that the proposed method can synthesize different facial expressions with kinds of poses of the face using a known facial expression and pose image.

Key words: tensor decomposition, multi-pose and expression synthesis, model subspace

中图分类号:

TP391.413

吕煊王志成赵卫东. 张量描述下的多姿态多表情人脸合成方法[J]. 计算机应用, 2012, 32(01): 256-260.

Lǚ Xuan WANG Zhi-cheng ZHAO Wei-dong. Multi-pose and expression face synthesis method based on tensor representation[J]. Journal of Computer Applications, 2012, 32(01): 256-260.

参考文献

[1]

ERSOTELOS N, FENG D. Building highly realistic facial modeling and animation: A survey ［J］. Visual Computer, 2008, 24(1): 13-30.

[2]

ZHANG QINGSHAN, LIU ZICHENG, GUO BAINING, et al. Geometry-driven photorealistic facial expression synthesis ［J］. IEEE Transactions on Visualization and Computer Graphics, 2006, 12(1): 48-60.

[3]

SIFAKIS E, NEVEROV I, FEDKIW R. Automatic determination of facial muscle activations from sparse motion capture marker data ［J］. ACM Transactions on Graphics, 2005, 24(3): 417-425.

[4]

WANG JUN, ZHANG CHANGSHUI, KOU ZHONGBAO. An analytical mapping for LLE and its applications in multi-pose face synthesis ［C］// Proceedings of the 14th British Machine Vision Conference. Norwich: British Machine Vision Association, 2003: 285-294.

[5]

VLASIC D, BRAND M, PFISTER H, et al. Face transfer with multilinear models ［J］. ACM Transactions on Graphics, 2005, 24(3): 426-433.

[6]

WANG HONGCHENG, AHUJA N. Facial expression decomposition ［C］// Proceedings of the Ninth IEEE International Conference on Computer Vision. Washington, DC: IEEE Computer Society, 2003: 958-965.

[7]

ALEX M, VASILESCU O, TERZOPOULOS D. Multilinear analysis of image ensembles: TensorFaces ［C］// ECCV'02: Proceedings of the 7th European Conference on Computer Vision.Berlin: Springer-Verlag, 2002: 447-460.

[8]

VASILESCU M A O, TERZOPOULOS D. Multilinear image analysis for facial recognition ［C］// Proceedings of the 16th International Conference on Pattern Recognition. Washington, DC: IEEE Computer Society, 2002: 511-514.

[9]

VASILESCU M A O, TERZOPOULOS D. Multilinear subspace a-nalysis of image ensembles ［C］// Proceedings of the 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC: IEEE Computer Society, 2003: 93-99.

[10]

LEE H S, KIM D. Tensor-based AAM with continuous variation estimation: Application to variation-robust face recognition ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(6): 1102-1116.

[11]

QIAO X, HAN S H, IGARASHI T, et al. Tensor-based subspace learning and its applications in multi-pose face synthesis ［J］. Neurocomputing, 2010, 73(13/14/15): 2727-2736.

[12]

KOLDA T G, BADER B W. Tensor decompositions and applications ［J］. SIAM Review, 2009, 51(3): 455-500.

[13]

COOTES T F, EDWARDS G J, TAYLOR C J. Active appearance models ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(6): 681-685.

[14]

The Weizmann face databases ［EB/OL］. [2011-04-06]. http://www.wisdom.weizmann.ac.il/~vision/FaceBase/.

张量描述下的多姿态多表情人脸合成方法

Multi-pose and expression face synthesis method based on tensor representation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 6

编辑推荐

Metrics

[1]	巩凯强, 张春梅, 曾光华. 卷积神经网络模型剪枝结合张量分解压缩方法[J]. 计算机应用, 2020, 40(11): 3146-3151.
[2]	任晓旭, 吕良福, 崔广泰. 耦合图像间的数据融合算法[J]. 计算机应用, 2018, 38(10): 2862-2868.
[3]	余可钦, 吴映波, 李顺, 蒋佳成, 向德, 王天慧. 基于移动用户上下文相似度的张量分解推荐算法[J]. 计算机应用, 2017, 37(9): 2531-2535.
[4]	丁小焕, 彭甫镕, 王琼, 陆建峰. 基于平行因子分解的协同聚类推荐算法[J]. 计算机应用, 2016, 36(6): 1594-1598.
[5]	丁小焕, 彭甫镕, 王琼, 陆建峰. 融合朋友关系和标签信息的张量分解推荐算法[J]. 计算机应用, 2015, 35(7): 1979-1983.
[6]	刘亚楠涂铮铮罗斌. 基于反馈稀疏约束的非负张量分解算法[J]. 计算机应用, 2013, 33(10): 2871-2873.