Abstract
Over the last few years, facial expression recognition is an active research field, which has an extensive range of applications in the area of social interaction, social intelligence, autism detection and Human-computer interaction. In this paper, a robust hybrid framework is presented to recognize the facial expressions, which enhances the efficiency and speed of recognition system by extracting significant features of a face. In the proposed framework, feature representation and extraction are done by using Local Binary Patterns (LBP) and Histogram of Oriented Gradients (HOG). Later, the dimensionalities of the obtained features are reduced using Compressive Sensing (CS) algorithm and classified using multiclass SVM classifier. We investigated the performance of the proposed hybrid framework on two public databases such as CK+ and JAFFE data sets. The investigational results show that the proposed hybrid framework is a promising framework for recognizing and identifying facial expressions with varying illuminations and poses in real time.
Highlights
Automatic facial expression recognition is the ability of the machine to automatically recognize expressions of emotions or expressions of social signals on faces
For feature extraction, we use integration of Local Binary Patterns (LBP), Histogram of Oriented Gradients (HOG) and compressive sensing. This stage can further divide into two steps: i) Feature vectors are extracted using LBP and HOG in both testing and training phases. ii) Compressive sensing technique is used for projection of extracted dense features vectors to lower space
We have randomly divided the dataset into 185 training images and 28 test images
Summary
Automatic facial expression recognition is the ability of the machine to automatically recognize expressions of emotions or expressions of social signals on faces. The three significant phases of the facial expression recognition system are 1) Face acquisition &pre-processing 2) Facial feature extraction and 3) Facial expression classification. Feature extraction for facial expression recognition broadly classified into three methods. There is a straight forward relation between features and distance between two facial points and the expressions. The appearance features based systems are based on the variations in the appearance of skin, furrows, wrinkles and texture changes In this method, features related to facial appearance changes of the whole face or specific face regions are extracted by applying image filters. Used for facial expression recognition, for example, Valstar et al presented an AAM which combines histogram with texture and shape parameters.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: International Journal of Engineering & Technology
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
