Mixture of experts for classification of gender, ethnic origin, and pose of human faces

被引:139
作者
Gutta, S [1 ]
Huang, JRJ
Jonathon, P
Wechsler, H
机构
[1] Philips Res Labs, Briarcliff Manor, NY 10510 USA
[2] George Mason Univ, Dept Comp Sci, Fairfax, VA 22030 USA
[3] Natl Inst Stand & Technol, Gaithersburg, MD 20899 USA
来源
IEEE TRANSACTIONS ON NEURAL NETWORKS | 2000年 / 11卷 / 04期
关键词
decision trees (DT); ensembles of radial basis functions (ERBF) ethnic classification; face processing; FERET; gating networks; gender classification; mixtures of experts; pose discrimination; radial basis functions (RBF's) kernels; support vector machines (SVMs);
D O I
10.1109/72.857774
中图分类号
TP18 [人工智能理论];
学科分类号
081104 [模式识别与智能系统]; 0812 [计算机科学与技术]; 0835 [软件工程]; 1405 [智能科学与技术];
摘要
In this paper we describe the application of mixtures of experts on gender and ethnic classification of human faces, and Dose classification, and show their feasibility on the FERET database of facial images. The FERET database allows us to demonstrate performance on hundreds or thousands of images. The mixture of experts is implemented using the "divide and conquer" modularity principle with respect to the granularity and/or the locality of information, The mixture of experts consists of ensembles of radial basis functions (RBFs), Inductive decision trees (DTs) and support vector machines (SVMs) implement the "gating network" components for deciding which of the experts should be used to determine the classification output and to restrict the support of the input space. Both the ensemble of RBF's (ERBF) and SVM use the RBF kernel ("expert") for gating the inputs. Our experimental results yield an average accuracy rate of 96% on gender classification and 92% on ethnic classification using the ERBF/DT approach from frontal face images, while the SVM yield 100% on pose classification.
引用
收藏
页码:948 / 960
页数:13
相关论文
共 48 条
[1]
MORE ABOUT THE DIFFERENCE BETWEEN MEN AND WOMEN - EVIDENCE FROM LINEAR NEURAL NETWORKS AND THE PRINCIPAL-COMPONENT APPROACH [J].
ABDI, H ;
VALENTIN, D ;
EDELMAN, B ;
OTOOLE, AJ .
PERCEPTION, 1995, 24 (05) :539-562
[2]
DEMOCRACY IN NEURAL NETS - VOTING SCHEMES FOR CLASSIFICATION [J].
BATTITI, R ;
COLLA, AM .
NEURAL NETWORKS, 1994, 7 (04) :691-707
[3]
An empirical comparison of voting classification algorithms: Bagging, boosting, and variants [J].
Bauer, E ;
Kohavi, R .
MACHINE LEARNING, 1999, 36 (1-2) :105-139
[4]
Bishop C. M., 1995, NEURAL NETWORKS PATT
[5]
Breiman L, 1998, ANN STAT, V26, P801
[6]
Bagging predictors [J].
Breiman, L .
MACHINE LEARNING, 1996, 24 (02) :123-140
[7]
SEX-DISCRIMINATION - HOW DO WE TELL THE DIFFERENCE BETWEEN MALE AND FEMALE FACES [J].
BRUCE, V ;
BURTON, AM ;
HANNA, E ;
HEALEY, P ;
MASON, O ;
COOMBES, A ;
FRIGHT, R ;
LINNEY, A .
PERCEPTION, 1993, 22 (02) :131-152
[8]
PARALLEL PROCESSING OF THE SEX AND FAMILIARITY OF FACES [J].
BRUCE, V ;
ELLIS, H ;
GIBLING, F ;
YOUNG, A .
CANADIAN JOURNAL OF PSYCHOLOGY-REVUE CANADIENNE DE PSYCHOLOGIE, 1987, 41 (04) :510-520
[9]
CARICATURAL EFFECTS IN AUTOMATED FACE PERCEPTION [J].
BRUNELLI, R ;
POGGIO, T .
BIOLOGICAL CYBERNETICS, 1993, 69 (03) :235-241
[10]
HUMAN AND MACHINE RECOGNITION OF FACES - A SURVEY [J].
CHELLAPPA, R ;
WILSON, CL ;
SIROHEY, S .
PROCEEDINGS OF THE IEEE, 1995, 83 (05) :705-740