Object tracking by energy maximization

P. Lappas

doi:10.1007/s00500-005-0459-y

Object tracking by energy maximization

2005-04-06

3

PRE

AI

DOI：10.1007/s00500-005-0459-y

Original

Original request for help

Save

Abstract

En 中文

We describe a robust algorithm for object tracking in long image sequences which extends the dynamic Hough transform to detect arbitrary shapes undergoing arbitrary affine motion. The proposed tracking algorithm processes the whole image sequence globally. First, the object boundary is represented in lookup-table form, and we then perform an operation that estimates the energy of the motion trajectory in the parameter space. We assign an extra term in our cost function to incorporate smoothness of changes due to rotation or scaling of the object. There is no need for training or initialization, and an efficient implementation can be achieved with coarse-to-fine dynamic programming and pruning. The method is shown to be robust under noise and occlusion and capable of tracking multiple objects.

Keywords:

pattern recognition

object tracking

hough transform

energy framework

Object tracking by energy maximization

Abstract

Journal

Organization