The integrated gaze and object tracking techniques to explo re the user's navigation

Abstract

This paper proposes a method to explore the user's navigation foci and visual tracks by estimating gaze points and mapping them to the objects of video content. The tracking method for multiple objects is derived from the adaptive weight based feature with probability densities. It is able to track the target objects efficiently even when the target objects are lost. It continuously applies sequence scheme and mean scheme throughout to track objects while they are lost. The experimental results demonstrate the proposed method provide higher robustness under different conditions.