Human expression recognition from motion using a radial basis function network architecture

被引:152
作者
Rosenblum, M
Yacoob, Y
Davis, LS
机构
[1] Computer Vision Laboratory, University of Maryland, College Park
来源
IEEE TRANSACTIONS ON NEURAL NETWORKS | 1996年 / 7卷 / 05期
关键词
D O I
10.1109/72.536309
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this paper a radial basis function network architecture is developed that learns the correlation of facial feature motion patterns and human expressions, We describe a hierarchical approach which at the highest level identifies expressions, at the mid level determines motion of facial features, and at the low level recovers motion directions, Individual expression networks were trained to recognize the ''smile'' and ''surprise'' expressions, Each expression network was trained by viewing a set of sequences of one expression for many subjects, The trained neural network was then tested for retention, extrapolation, and rejection ability, Success rates were 88% for retention, 88% for extrapolation, and 83% for rejection.
引用
收藏
页码:1121 / 1138
页数:18
相关论文
共 34 条
[11]  
Ekman P., 1975, UNMASKING FACE
[12]  
ESSA IA, 1994, 1994 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS, P76, DOI 10.1109/CVPR.1994.323813
[13]  
Goldstein E. B., 1989, Sensation and perception
[14]  
GRIMSON WEL, 1992, P DARPA IM UND WKSHP
[15]  
Hertz J., 1991, Introduction to the Theory of Neural Computation
[16]  
HOWELL JA, 1989, P INT C ACC LARG EXP
[17]  
JONES RD, 1990, P INT JOINT C NEUR N
[18]  
JONES RD, 1990, P NONL CHAOT PROC PL
[19]  
JONES RD, 1990, P 1 LOS AL WKSHP COG
[20]  
LEE YC, 1989, P 13 C NUM SIM PLASM