Object Tracking Algorithm based on Bilateral Structure Tensor under Lie Group Manifold
Download as PDF
DOI: 10.25236/meici.2019.061
Author(s)
Yinghong Xie, Qiang Guo, and Chunhe Shi
Corresponding Author
Yinghong Xie
Abstract
Considering of object modeling problem with the complex background, the paper proposes a new object tracking algorithm based on bilateral structure tensor using particle filter under Lie group. Bilateral structure tensor is adopted to describe the appearance of the object, which can maintain the information of image edge better. At the same time, for the bilateral structure obeys Lie group structure, Riemann geometry which is much more accurate is applied to calculate the mean value and design particle filtering algorithm. The experimental results show that the proposed algorithm can achieve stable tracking for the visual and the variable targets under water
Keywords
Object tracking, Bilateral structure tensor, Particle filtering, Lie group, Riemann geometry