Head tracking by means of probabilistic neural networks

In this paper, a neural network based method is applied to track a moving person's head in image sequences. The head's colour and shape are chosen manually as prototypes, when the first frame becomes available. Then estimations of the head's position are made in the following frame. After that, probabilistic neural networks (PNN) are used to classify these estimations into two categories: head and background. Estimations belonging to the first category are then put through a competitive layer. The winner is then regarded as the optimal estimate for the position of the head in that frame. The tracking involves finding the optimal position frame by frame. A simple data fusion algorithm is also applied to improve the performance of neural networks. With the help of PNN, the quantity of estimations that needs to be processed is decreased significantly. This approach can deal with tolerant rotation and occlusions in a complex background.

[1]  Michael J. Black.  Recursive Non-Linear Estimation of Discontinuous Flow Fields , 1994, ECCV.

[2]  Shang-Liang Chen,et al.  Orthogonal least squares learning algorithm for radial basis function networks , 1991, IEEE Trans. Neural Networks.

[3]  Patrick Pérez,et al.  Data fusion for visual tracking with particles , 2004, Proceedings of the IEEE.

[4]  Jochen Triesch,et al.  Democratic Integration: Self-Organized Integration of Adaptive Cues , 2001, Neural Computation.

[5]  Stanley T. Birchfield,et al.  Elliptical head tracking using intensity gradients and color histograms , 1998, Proceedings. 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No.98CB36231).

[6]  Bernt Schiele,et al.  Towards robust multi-cue integration for visual tracking , 2001, Machine Vision and Applications.

[7]  Meng Joo Er,et al.  High-speed face recognition based on discrete cosine transform and RBF neural networks , 2005, IEEE Transactions on Neural Networks.

[8]  Lin Guo,et al.  Human face recognition based on radial basis probabilistic neural network , 2003, Proceedings of the International Joint Conference on Neural Networks, 2003..

[9]  Philip D. Wasserman,et al.  Advanced methods in neural computing , 1993, VNR computer library.

[10]  Weifeng Tian,et al.  Head tracking based on the integration of two different particle filters , 2006 .

[11]  Martin T. Hagan,et al.  Neural network design , 1995 .

[12]  S. Birchfield,et al.  An elliptical head tracker , 1997, Conference Record of the Thirty-First Asilomar Conference on Signals, Systems and Computers (Cat. No.97CB36136).

[13]  Bohyung Han,et al.  Bayesian Filtering and Integral Image for Visual Tracking , 2005 .

[14]  Anton van den Hengel,et al.  Probabilistic Multiple Cue Integration for Particle Filter Based Tracking , 2003, DICTA.

[15]  Luc Van Gool,et al.  An adaptive color-based particle filter , 2003, Image Vis. Comput..

[16]  Meng Joo Er,et al.  Face recognition with radial basis function (RBF) neural networks , 2002, IEEE Trans. Neural Networks.

[17]  Ying Wu,et al.  Nonstationary color tracking for vision-based human-computer interaction , 2002, IEEE Trans. Neural Networks.

[18]  Fan Yang,et al.  Implementation of an RBF neural network on embedded systems: real-time face tracking and identity verification , 2003, IEEE Trans. Neural Networks.

[19]  Rama Chellappa,et al.  Visual tracking and recognition using appearance-adaptive models in particle filters , 2004, IEEE Transactions on Image Processing.