Pose robust face recognition based on CASPCM model

ZHAO Ming-hua; YOU Zhi-sheng; YU Jing; XIONG Yun-yu

CASPCM model is proposed to make up the disadvantages of ASPCM model while dealing with faces with large angles.The training samples are grouped according to their distances to model centers and a local ASPCM model is constructed for each group.Synthesis result of CASPCM model is obtained by averaging results of the local ASPCM models with appropriate weights.Gradient-descent algorithm is used to iteratively improve estimate of the head pose.Accuracy and generalization are used to gauge analysis and synthesis abilities of the model.Experimental results show that the two abilities of CASPCM model are both superior to ASPCM model;recognition ratio of CASPCM model is 7% higher than ASPCM model.

Corresponding author: YOU Zhi-sheng, zsyou@mail.sc.cninfo.net;

1. Institute of Image & Graphic, College of Computer Science, Sichuan University, Chengdu 610064, China

Received Date: 2005-05-23

Accepted Date: 2005-07-21

Available Online: 2006-07-25

Abstract: CASPCM model is proposed to make up the disadvantages of ASPCM model while dealing with faces with large angles.The training samples are grouped according to their distances to model centers and a local ASPCM model is constructed for each group.Synthesis result of CASPCM model is obtained by averaging results of the local ASPCM models with appropriate weights.Gradient-descent algorithm is used to iteratively improve estimate of the head pose.Accuracy and generalization are used to gauge analysis and synthesis abilities of the model.Experimental results show that the two abilities of CASPCM model are both superior to ASPCM model;recognition ratio of CASPCM model is 7% higher than ASPCM model.

HTML

Reference (10)

[1]	LIU Zh F,YOU Zh S,WANG Y Q.Face recognition based on PCA and ICA[J].Laser Technology,2004,28(1):78～81(in Chinese).
[2]	KOUZANI A Z,HE F,SAMMUT K.Towards invariant face recognition[J].Information Sciences,2000,123(1):75～101.
[3]	LEE M W,SURENDRA R.Pose-invariant face recognition using a 3-D deformable model[J].Pattern Recognition,2003,36(8):1835～1846.
[4]	LAI J H,YUEN P C,FENG G C.Face recognition using holistic Fourier invariant features[J].Pattern Recognition,2001,34(1):95～109.
[5]	EICKELER S,MULLER S,RIGOLL G.Recognition of JPEG com-pressed face images based on statistical methods[J].Image and Vision Computing,2001,18(4):279～287.
[6]	OKADA K,AKAMATSU S,VON D M C.Analysis and synthesis of pose variations of human faces by alinear PCMAP model and its application for pose-invariant face recognition system[A].Proceedings Fourth IEEE International Conference on Automatic Face and Gesture Recognition[C].USA:IEEE Computer Society,2000.142～149.
[7]	TROGE N F,BULTHOFF H.Face recognition under varying poses:the role of texture and shape[J].Vision Research,1996,36(12):1761～1771.
[8]	RICHARD O,PETER E,DAVID G.Pattern classification[M].2nd ed,Beijing:China Machine Press,2003.456～457(in Chinese).
[9]	BIAN Z Q,ZHANG X G.Pattern recognition[M].2nd ed,Beijing:Tsinghua University Press,2000.331～332(in Chinese).
[10]	TRESP V,HOFMANN R.Nonlinear time-series prediction with missing and noisy data[J].Neural Computation,1998,10(3):731～747.

Pose robust face recognition based on CASPCM model

Abstract

References

Proportional views

通讯作者: 陈斌, bchen63@163.com

Proportional views