Abstract: Automatic facial expression analysis is one of the an interesting and challenging problem and also impacts important applications in many areas such as human–computer interaction and data-driven animation. An important step for successful facial expression recognition is to deriving an effective facial representation from original face images. For this, empirically evaluate facial representation majorly based on statistical local features, Local Binary Patterns for person-independent facial expression recognition. Several machine learning methods are systematically examined on several databases. Local Binary Pattern features are effective and much more efficient for facial expression recognition. And formulating Boosted-LBP further to extract the most discriminant LBP features and the best recognition performance is obtained by using Support Vector Machine classifiers with Boosted LBP features. An investigation on LBP features for low-resolution facial expression recognition which is a critical problem but seldom addressed in the existing work. According to observation of experiments that LBP features perform not only stably but also robustly over a useful range of low resolutions of face images and yield promising performance in compressed low-resolution video sequences captured in real-world environments. Facial expressions are one of the most critical sources of variation in face recognition, especially in the frequent case where only a single sample per person is available for enrollment. Some methods that improve the accuracy in the presence of such variations are still required for a reliable authentication system. Because of this, we address the problem with an analysis by- synthesis-based scheme in which a number of synthetic face images with different expressions are produced. For this an animatable 3D model is generated for each user based on 17 automatically located landmark points and the contribution of these additional images in terms of the recognition performance is evaluated with three different techniques such as principal component analysis(PCA), Linear Discriminant Analysis(LDA) and local binary patterns(LBP) on face recognition. Significant improvements are achieved in face recognition accuracies for each algorithm.

Keywords: LDA, LBP, PCA