Particle Swarm Optimization algorithm for facial emotion detection

Particle Swarm Optimization (PSO) algorithm has been applied and found to be efficient in many searching and optimization related applications. In this paper, we present a modified version of the algorithm that we successfully applied to facial emotion detection. Our approach is based on tracking the movements of facial action units (AUs) placed on the face of a subject and captured in video clips. We defined particles that form swarms such that they have a component around the neighborhood of each AU. Particles are allowed to move around the effectively n-dimensional search space in search of the emotion being expressed in each frame of a video clip (where n is the number of action units being tracked). We have implemented and tested the algorithm on video clips that contain three of the six basic emotions, namely happy, sad and surprise. Our results show the algorithm to have a promising success rate.