Abstract Deriving an effective facial representation from original face images is a vital step for successful facial expression recognition. In this paper, we explore the effect of representing the information in facial expression recognition using local patch local binary pattern (LBP). The processed information with accurate representation of different expressions can discriminate and improve the overall facial expression recognition accuracy. The objective of this paper is to analyze the effect of using local region patch to represent facial features with local binary pattern as the feature extraction method. In our experiment, first facial landmark is being detected using cascade linear regression, followed by alignment and normalization. The LBP feature extraction is being performed on the holistic image, followed by patch of mouth and eyes. At the final stage support vector machine (SVM) has been used as a classifier to examine the recognition rate. The results are being validated using the extended Cohn-Kanade database. From the analysis, it is found that using local region LBP can significantly reduce the number of features to be fed into SVM. Hence the processing time is improved.
Analysis of Local Binary Pattern for Facial Expression Recognition Using Patch Local Binary Pattern on Extended Cohn Kanade Database
10th International Conference on Robotics, Vision, Signal Processing and Power Applications ; 10 ; 633-639
Lecture Notes in Electrical Engineering ; 547 , 10
01.01.2019
7 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Robust Facial Expression Recognition using Local Binary Patterns
British Library Conference Proceedings | 2005
|Discriminative local binary pattern
British Library Online Contents | 2016
|Facial expression recognition based on local binary patterns
British Library Online Contents | 2007
|