Robust real-time periodic motion detection, analysis, and applications

We describe new techniques to detect and analyze periodic motion as seen fr om both a static and a moving camera. By tracking objects of interest, we c ompute an objects self-similarity as it evolves in time. For periodic motio n, the self-similarity measure is also periodic and we apply Time-Frequency analysis to detect and characterize the periodic motion. The periodicity i s also analyzed robustly using the 2D lattice structures inherent in simila rity matrices. A real-time system has been implemented to track and classif y objects using periodicity. Examples of object classification (people, run ning dogs, vehicles), person counting, and nonstationary periodicity are pr ovided.