Visual tracking with a cognitive observation model
Xi Chen, Yuan Zhang, Zhaohui Li · 2016
In this paper, we present a visual mechanism and a visual saliency model which is suitable for the visual tracking algorithm. In order to extract robust feature of motion area, instead of using saliency map to detect moving targets, our algorithm adopts a bottom-up attention model based on the human visual information processing mechanism. The method is robust to illumination and viewpoint changes and applicable to indoor as well as outdoor scenes. In addition, experiments were conducted to compare system performances between the proposed algorithms and the Camshift algorithm.