Goals:
Presenting a fast algorithm for tracking object of interest in a sequence of images based on multiresolution processing methods. The presented algorithm has to be implemented in real-time.

Project Summary:
Object of interest in the consecutive frames of a video may move, rotate, shrink, and enlarge. Multiresolution based processing methods are one of the best choices for overcoming tracking problems in those circumstances. Feature extraction in wavelet domain is the main part for most of the successful multiresolution based algorithms. Following our previous work [AmiRBK03], we will try to take advantage of the new Integer Lifting scheme Wavelet Transform (ILWT). Using wavelet Packet Transform (WPT) we had spared just enough time to do a simple processing in wavelet domain at [AmiRBK03]. Then partial occlusion had severe damage to successful tracking. ILWT is much faster than WPT so having more complex processing in wavelet domain, we hope to have successful object tracking in cluttered scenes.

Team Members:
Mahdi Amiri