On-line Evolutionary Head Pose Measurement by Feedforward Stereo Model Matching

This paper presents a method to estimate the 3D pose of a human's head using two images input from stereo cameras. The proposed method utilizes an evolutionary search technique of genetic algorithm (GA) and a fitness evaluation based on a stereo model matching. To improve the dynamics of recognition, a motion-feedforward method is proposed for the hand-eye system. The effectiveness of the method is confirmed by the experiments where the motion of the hand-eye camera compensated for the relative motion of the object in camera frame, resulting robust recognition against the hand-eye motion

[1]  Vincent Lepetit,et al.  Stable real-time 3D tracking using online and offline information , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  H. Suzuki,et al.  Visual servoing to catch fish using global/local GA search , 2005, IEEE/ASME Transactions on Mechatronics.

[3]  Hongbin Zha,et al.  Cooperative manipulations based on genetic algorithms using contact information , 1995, Proceedings 1995 IEEE/RSJ International Conference on Intelligent Robots and Systems. Human Robot Interaction and Cooperative Robots.

[4]  Ruigang Yang,et al.  Model-based head pose tracking with stereovision , 2002, Proceedings of Fifth IEEE International Conference on Automatic Face Gesture Recognition.

[5]  Mamoru Minami,et al.  Real-time face detection using hybrid GA based on selective attention , 2004, 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566).

[6]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[7]  Michel Dhome,et al.  Real time 3D template matching , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[8]  Roberto Cipolla,et al.  Determining the gaze of faces in images , 1994, Image Vis. Comput..