Real-Time Tracking Using Wavelet Representation

We have recently developed an original tracking framework allowing to track textured templates in real time [3]. This framework is based on the use of difference images, the difference between the target template and the template included in predicted area of interest. For efficiency purposes, the difference image was limited to a few points belonging to the area of interest. The measurements were therefore very punctual. In this article a wavelet representation of the area of interest is used as a substitute for this punctual measurement. In this case, the difference image is a difference of wavelet parameters (difference between the wavelet representation of the target and the wavelet representation of the template included in the current position of the region of interest). This algorithm is a part of a real-time system aiming at automatically detecting and tracking vehicles in video sequences.

[1]  Chris J. Harris,et al.  VEHICLE DETECTION AND RECOGNITION IN GREYSCALE IMAGERY , 1995 .

[2]  F. Chausse,et al.  A fast and robust vision based road following algorithm , 2000, Proceedings of the IEEE Intelligent Vehicles Symposium 2000 (Cat. No.00TH8511).

[3]  Tomaso A. Poggio,et al.  A general framework for object detection , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[4]  Paul A. Viola,et al.  Robust Real-Time Face Detection , 2001, International Journal of Computer Vision.

[5]  P. Sayd.  A DYNAMIC VISION ALGORITHM TO RECOVER THE 3D SHAPE OF A NON-STRUCTURED ROAD , 1998 .

[6]  Michel Dhome,et al.  A simple and efficient template matching algorithm , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[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.