Title :
Visual object tracking based on filtering methods
Author :
Wang, Kun ; Liu, Xiaoping P.
Author_Institution :
Dept. of Syst. & Comput. Eng., Carleton Univ., Ottawa, ON, Canada
Abstract :
Visual object tracking is an important, but open research topic in many practical applications. In this paper, the particle filter, a filtering algorithm based on the sequential-importance-sampling (SIS), is developed and implemented with different modifications to the transition models and constraint conditions. By applying the particle filter into a typical object tracking task, several experimental results are obtained and the feasibility of the modified particle filter is verified.
Keywords :
object tracking; particle filtering (numerical methods); filtering algorithm; filtering methods; particle filter; sequential importance sampling; visual object tracking; Image edge detection; Particle filters; Proposals; Target tracking; Video sequences; filtering and data association; particle filter; system transition model; visual object tracking;
Conference_Titel :
Instrumentation and Measurement Technology Conference (I2MTC), 2011 IEEE
Conference_Location :
Binjiang
Print_ISBN :
978-1-4244-7933-7
DOI :
10.1109/IMTC.2011.5944102
