Title :
Tracking in clutter based on Mean Shift embedded particle filter
Author :
Zheng, Lin ; Liu, Quan
Author_Institution :
Sch. of Inf. Eng., Wuhan Univ. of Technol., Wuhan, China
Abstract :
In this paper, we present a new Mean Shift embedded particle filter for visual tracking. Two kinds of Mean Shifts are used. The pixel based Mean Shift is employed to optimize each particle independently and locally. Then the particle based Mean Shift is employed to optimize all the particles dependently. This algorithm is used to track objects in the cluttered environment. The experiments show that the method performs robust in complex situation.
Keywords :
Monte Carlo methods; clutter; object detection; particle filtering (numerical methods); clutter tracking; mean shift embedded particle filter; object tracking; sequential Monte Carlo techniques; visual tracking; Filtering theory; Information filtering; Information filters; Monte Carlo methods; Optimization methods; Particle filters; Particle tracking; Probability density function; Robustness; Sliding mode control; Particle Filter; Particle based Mean Shift; Pixel based Mean Shift;
Conference_Titel :
Computer Engineering and Technology (ICCET), 2010 2nd International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-6347-3
DOI :
10.1109/ICCET.2010.5486215
