Feature Extraction Techniques for Facial Expression Recognition

The area of automatic Facial Expression recognition has received an increasing attention from the researchers over last two decades due to its applications in various areas like human-computer interaction systems, Human Machine Intelligent Interaction (HMII), intelligent robotics, human centered transportation system for driver’s fatigue measurement, tools for people with lower vision and health care robots etc. An efficient feature extraction technique is an essential component of an automatic FER system. The feature extraction techniques widely used are categorized into: Appearance based methods and Geometric methods. Several works presented by the researchers use a hybrid approach combining both methods to create the feature vector for classification. In this paper, a survey is presented on different types of feature extraction techniques.