Variational Shape Model for Lip Postures Recognition Using GA

A system for feature extraction and recognition of lip postures is presented, by constructing a shape model of the inner and outer contours of the lips, whose parameters are controlled by a canonical genetic algorithm. Features of each posture are the model parameters which better fit to the analyzed image. A Bayesian classifier is used to classify, under the assumption that there is equiprobability between the classes, obtaining results up to 82.4% of accurate classification