Eye tracking plays an important role in understanding human visual perception, especially visual attention and quality assessment. In this paper, we propose an efficient algorithm for identifying the region of attention in the eye tracking protocol. We use velocity-threshold based fixation identification algorithm to identify fixations and divide them into different clusters. In a particular cluster, we propose a densest position based method to locate the center of human attention. We have implemented the eye tracking protocols and conducted the experiments to evaluate the performance of our system. As shown in our results, the proposed method shows superior performance when compared to the traditional method.