Accurate real-time object tracking with linear prediction method

This paper presents an efficient technique for real-time tracking of a single moving object in terrestrial scenes using a stationary camera. The tracking algorithm is based on the linear prediction (LP) solved by the maximum entropy method (MEM). It attempts to predict the centroid of the moving object in the next frame, based on several past centroid measurements. Using a second order of the linear prediction method, the proposed algorithm is able to accurately track the moving object. It is shown analytically that the proposed recursive predictor-corrector tracking algorithm is able to yield high accuracy performance and is superior to that of the Kalman filter, for a possibly random movement of single moving object.