Facial feature extraction using hybrid genetic-simplex optimization in multi-objective Active Appearance Model

This paper presents an optimization methodology of hybrid multi-objective genetic algorithm for the real world optimization problem of facial analysis of multiple camera images by 2.5D Active Appearance Model (AAM). Facial large lateral movements make acquisition and analysis of facial images by single camera inefficient. Moreover non-convex multidimensional facial search space formed by AAM requires an efficient optimization methodology. Currently with wide availability of inexpensive cameras a multi-view system is as practical as single-view. To manage these multiple information multi-objective genetic algorithm is employed for optimizing the face search. To efficiently tackle the problem of non-convexity of the search space, hybridization of NSGA-II (Non-dominated Sorting Genetic Algorithm) with Nelder Mead Simplex (HMGSO-AAM) is proposed in this paper. For this hybridization we develop a unique method of calculating the relevant information of each camera in a multiple camera system which makes hybridization of Simplex with NSGA-II efficient and robust. Our proposed algorithm is applied on different facial poses of CMU-PIE database, real webcam images and synthetic facial images. Simulation results validate the efficiency, accuracy and robustness achieved.