Evolutionary head pose measurement by stereo model matching

This paper presents a method to estimate the 3D pose of a human's head using two images input from two stereo cameras. The proposed method utilizes an evolutionary search technique of genetic algorithm (GA) and a fitness evaluation based on a matching stereo model. Here, the head position is detected by an optimization of the computation of the brightness difference between an internal surface of head and a contour-strips, meanwhile, the head orientation is measured using color information about the skin region and the hair region of the input images. Moreover, the human's eyes, as one of the facial features are also recognized to improve the robustness of head pose measurement. Experiments to estimate three different pose of an arbitrary person's head have been conducted to show the effectiveness of the proposed method.