The recognition of basic facial expressions by neural network

Author

Kobayashi, Hiroshi ; Hara, Fumio

Author_Institution

Dept. of Mech. Eng. Sci., Tokyo Univ., Japan

fYear

1991

fDate

18-21 Nov 1991

Firstpage

460

Abstract

An attempt is being made to develop a mind-implemented robot that can carry out intellectual conversation with humans. As the first step for this study, a method for the robot to perceive human emotion is investigated. Specifically, a method of human emotion recognition from facial expressions by a neural network is shown. The authors categorized facial expressions into six groups (surprise, fear, disgust, anger, happiness, and sadness) and obtained 70 CCD (charge coupled device) camera-acquired data of facial feature-points from three components of the face (eyebrows, eyes, and mouth). Then the facial expression information is generated and input into the input units of the neural network; network learning and recognition are done by a backpropagation algorithm

Keywords

computer vision; computerised pattern recognition; learning systems; neural nets; CCD camera; backpropagation algorithm; facial expression recognition; facial feature-points; human emotion recognition; learning systems; neural network; pattern recognition; robot vision; Charge coupled devices; Charge-coupled image sensors; Emotion recognition; Eyebrows; Eyes; Face recognition; Humans; Mouth; Neural networks; Robots;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1991. 1991 IEEE International Joint Conference on

Print_ISBN

0-7803-0227-3

Type

conf

DOI

10.1109/IJCNN.1991.170444

Filename

170444