Title :
Efficient Multi-target Human Motion Tracking
Author :
Abhyankar, Aditya
Abstract :
Effective human motion tracking is necessary for all surveillance based biometric applications. Human motion being highly non-linear and non-modular is always difficult to track. To make it more challenging, in real scenarios, there are always multiple targets to be tracked. In this work particle filter based multiple target tracking system is proposed. For performing actual tracking Markov chain Monte Carlo methods are used. The system was tested using moviedb and labdb databases and accuracy of 91.7% was obtained.
Keywords :
Markov processes; Monte Carlo methods; biometrics (access control); image motion analysis; particle filtering (numerical methods); surveillance; target tracking; Markov chain Monte Carlo method; labdb database; moviedb database; multitarget human motion tracking; particle filter; surveillance based biometric application; Biometrics; Humans; Integral equations; Motion analysis; Particle filters; Particle tracking; Springs; State-space methods; Surveillance; Target tracking; Markov Chains; Monte Carlo method; Motion Tracking; Multiple Targets; Particle filtering;
Conference_Titel :
Computer Science and Information Technology - Spring Conference, 2009. IACSITSC '09. International Association of
Conference_Location :
Singapore
Print_ISBN :
978-0-7695-3653-8
DOI :
10.1109/IACSIT-SC.2009.100
