Facial expression recognition with LBP and SLPP combined method

Author

Dan Han ; Yue Ming

Author_Institution

Beijing Key Lab. of Work Safety Intell. Monitoring, Beijing Univ. of Posts & Telecommun., Beijing, China

fYear

2014

fDate

19-23 Oct. 2014

Firstpage

1418

Lastpage

1422

Abstract

The way to extract facial expression features from 3D face image is significant in 3D expression recognition. However, most of the feather extraction methods are based on geometric. In this paper, we extend a combined strategy for Local Binary Patterns (LBP) and Supervised Locality Preserving Projection (SLPP) in facial expression recognition. First, we use LBP to get the histogram of the image. Then the SLPP method is used to reduce the dimension. At last, the k-Nearest Neighbor algorithm is used as the classifier. The recognition rate of the combined method is compared with the traditional methods including PCA, LDA and SLPP. The result shows that the recognition accuracy of the LBP and SLPP combined method is better than that the others in the field of 3D facial expression recognition..

Keywords

emotion recognition; face recognition; feature extraction; image classification; pattern clustering; 3D face image; 3D facial expression recognition; LBP method; SLPP method; facial expression feature extraction; image histogram; k-nearest neighbor classifier algorithm; local binary pattern strategy; supervised locality preserving projection strategy; Databases; Face; Face recognition; Feature extraction; Image recognition; Principal component analysis; Three-dimensional displays; Facial Expression Recognition; K-Nearest Neighbors Classifier; Local Binary Patterns; Supervised Locality Preserving Projection;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing (ICSP), 2014 12th International Conference on

Conference_Location

Hangzhou

ISSN

2164-5221

Print_ISBN

978-1-4799-2188-1

Type

conf

DOI

10.1109/ICOSP.2014.7015233

Filename

7015233