Facial Expression Recognition using Anatomy Based Facial Graph

Automatic analysis of human facial emotions is one of the challenging problems in intelligent systems and social signal processing. It has many applications in human-computer interactions, social robots, interactive multimedia and behavior monitoring. In this paper, our specific aim is to develop a method for facial movement recognition based on verifying movable facial elements and estimate the movements after any facial expressions. The algorithm plots a face model graph based on facial expression muscles in each frame and extracts features by measuring facial graph edges' size and angle variations. Seven facial expressions, including neutral pose are being classified in this study using support vector machine and other classifiers on MMI databases. The approach does not rely on action unit system, and therefore eliminates errors which are otherwise propagated to the final result due to incorrect initial identification of action units. Experimental results show that analyzing facial movements gives accurate and efficient information in order to identify different facial expressions.