Face Pose Estimation by Chaotic Artificial Bee Colony

This paper focuses on the problem of human face pose estimation using single image. Traditional methods for 2D-3D feature based pose estimation problem require two inputs, and they can not work well due to lack of correspondences of input images. In order to solve the problem more effectively and rapidly, a six-point template was used which overcomes the shortcomings of traditional approaches. Besides, a novel chaotic artificial bee colony based on artificial bee colony and Rossler attractor was proposed to solve the six-point template. Experiments on 40 different poses demonstrate that the proposed method is more accurate, robust and antinoising than both genetic algorithm and particle swarm optimization.