Geometric and Appearance Feature Analysis for Facial Expression Recognition

This paper evaluates facial recognition based on Local Binary Patterns on three orthogonal planes, Pyramid of Histogram of Gradients, and a geometrical feature based on distance between fiducial points for person-independent facial expression recognition. Different machine learning methods are systematically examined on two databases (posed and spontaneous). For posed database Cohn-Kanade has been used and for spontaneous database FeedTUM has been used. Extensive experiments illustrate that Local Binary Patterns on three orthogonal planes is effective and efficient for facial expression recognition. The best recognition performance is obtained by using Support Vector Machine classifier.