Robust object tracking using kernalized correlation filters (KCF) and Kalman predictive estimates

Abstract

Visual object tracking and detection is an advanced interdisciplinary research area which is crucial for many surveillance security applications. In this paper, we aim to track moving objects more accurately and significantly faster when compared to other approaches. This can be achieved through Kernalized Correlation Filters (KCF). The proposed work adopts a novel approach where the KCF filter is enhanced by integrating it with Kalman filter. The integrated Kalman based KCF (KKCF) tracker outperforms the traditional KCF by performing well for outlier or failure cases which is corrected through Kalman filter. Experimental results show the performance compared to KCF and other existing methods.