Facial expression recognition using high order directional derivative local binary patterns

Abstract

The most expressive manner which human can reveal his emotional states is facial expression. Automatic facial expression recognition is an emerging field of study having extensive applications among which the human-computer interaction (HCI) has received lots of attentions in recent years. The features extracted from facial images, in order to recognize facial expressions, play an essential role in effectiveness of the facial image descriptors. Local binary pattern (LBP) texture descriptors have been known as simple, yet efficient descriptors which are noticeably used for extracting facial patterns from images. Recently, a generalized form of local binary pattern has been introduced which can offer a more precise image description than simple LBP descriptors. Consequently, it would be expected that taking the advantage of using these new LBP texture descriptors will produce more promising results in comparison with use of simple local binary pattern descriptors. In this paper, a novel method has been proposed for image feature extraction using these new image texture descriptors (generalized LBP); then, the obtained results have been compared to the results produced when applying simple LBP descriptors. Furthermore, K-NN and SVM have been used as classifiers in the proposed approach. Finally, a comparison between the proposed method and the existing local binary pattern algorithms for facial expression recognition concludes the superiority of the proposed algorithm over its existing counterparts. Keywords-derivative local binary patterns, facial expression recognition, support vector machine.

Facial Expression Recognition by Jointly Partial Image and Deep Metric Learning

“…In addition to the abovementioned methods used to extract the 2D features of static images, focusing on the The associate editor coordinating the review of this manuscript and approving it for publication was Yudong Zhang . temporal and spatial information in an image sequence method is also proposed, such as using spatiotemporal covariance descriptors (Cov3D) [8], the temporal modeling of the shape (TMS) [9], expressionlets on a spatiotemporal manifold (STM-ExpLet) [10], etc. The FER method based on human-crafted features requires additional classifiers for classification, such as K-NN classifier [11], the SVM classifier [12], and the Hidden Markov model [13]. Although this method has been applied in more cases, its features tend to be relatively singular, and they are susceptible to disruptions caused by head pose and illumination changes [14].…”