Analysis and Study of Visual Tracking Algorithms

Mahdi Tanbakuchi¹ Mojtaba Lotfizad²

1) Student Department of Electrical and Computer Engineering South Tehran Branch Islamic Azad University Tehran, Iran Email:
2) Assistant Professor Department of Electrical and Computer Engineering Tarbiat Modares University Tehran, Iran Email:

Publication : International Conference on Science and Engineering(2icesconf.com)
Abstract :
In this paper, we analyze the most recent visual tracking algorithms. By studying the building blocks of such algorithms, we realize that these algorithms are based on either detection or stochastic filtering. Both methods need a way of modelling the target so that it can be recognized in subsequent frames; therefore, having algorithms that can describe the targets is crucial. Since the accuracy of the tracking algorithm relies on the descriptor, some competent descriptors are introduced here. To better understand the details of each type of algorithms, some of the most recent visual trackers are studied and their performances are compared.
Keywords : feature extraction descriptor visual tracking classification particle filter