全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于特征运动的表情人脸识别

DOI: 10.11834/jig.2002011340

Keywords: 特征运动,人脸识别,运动向量残差,运动空间,面部表情识别,模式识别

Full-Text   Cite this paper   Add to My Lib

Abstract:

人脸像的面部表情识别一直是人脸识别的一个难点,为了提高表情人脸识别的鲁棒性,提出了一种基于特征运动的人脸识别方法,该方法首先利用块匹配的方法来确定表情人脸和无表情人脸之间的运动向量,然后利用主成分分析方法(PCA)从这些运动向量中,产生低维子空间,称之为特征运动空间,测试时,先将测试人脸与无表情人脸之间的运动向量投影到特征运动空间,再根据这个运动向量在特征运动空间里的残差进行人脸识别,同时还介绍了基于特征运动的个人模型方法和公共模型方法,实验结果证明,该新算法在表情人脸的识别上,优于特征脸方法,有非常高的识别率。

References

[1]  [1]Chellappa R, Wilson C L, Sirohey S. Human and machine recognition of faces: a survey [J]. Proceedings of the IEEE,1995,83(5) :705~741.
[2]  [3]Pentland A, Moghaddam B, Starner T. View-based and modular eigenspaces for face recognition[A]. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition 1994 [C], Seattle, WA, USA, 1994:84~91.
[3]  [8]Tekalp A M. Digital video processing[M]. Englewood Cliffs,NJ: Published by Prentice Hall, Inc., a Simon & Schuster Company, 1995:95~116.
[4]  [2]Turk M, Pentland A. Eigenfaces for recognition[J]. Journal of Cognitive Neuroscience, 1991,3(1):71~86.
[5]  [4]Gao Y, Leung M K H, Wang W et al. Fast face identification under varying pose from a single 2-D model view [J].Proceedings of the IEEE, Vision, Image and Signal Processing,2001,148(4): 248~253.
[6]  [5]Lien J J, Kanade T, Cohn J F et al. Automated facial expression recognition based on FACS action units[A]. In: Proceedings of Third IEEE International Conference on Automatic Face and Gesture Recognition[C], Nara, Japan. 1998:390~395.
[7]  [6]Yacoob Y, Lam H, Davis L S. Recognizing faces showing expressions [A ]. In: International Workshop on Automatic Face-and Gesture-Recognition[C], Zurich, 1995: 278~283.
[8]  Belhumeur P N, Hespanha J P, Kriegman D J. Eigenfaces vs.Fisherfaces: Recognition using class specific linear projection [J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 1997,19(7) : 711~720.
[9]  [9]Carnegie Mellon University face expression database[DB/OL]:http ://amp. ece. cmu. edu/downloads. btm.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133