Robust Object Tracking Based on a Novel Feature

Abstract

This paper proposes a powerful and robust local descriptor, called color Weber feature(CWF). The CWF descriptor consists of two components: color contrast ratio and color edge orientation. Inspired by Weber's Law, we propose color contrast ratio which implements hierarchical quantization of salience within an image to simulate the pattern perception of human beings. We embed the proposed CWF representation model in the mean shift tracking framework to perform object tracking. The experiments results demonstrate that CWF is a viable object representation for tracking even in the adverse scenarios such as heavy occlusions, illumination variations and similar objects interference.