A novel adaptive object tracking method based on Expected Likelihood Kernel

摘要

Visual tracking is a new line of broad research. It is required for advanced vision-based applications such as visual surveillance and vision-based human-robot interaction. In this paper, we propose a new method of object detection and tracking algorithm using Adaptive Expected Likelihood Kernel. In this algorithm we combine between the probability product kernels as a similarity measure, and the integral image to increase the speed of the algorithm.