Comparison of visual tracking algorithms on in vivo sequences for robot-assisted flexible endoscopic surgery

Flexible endoscopes are used in many diagnostic and interventional procedures. Physiological motions may render the physician's task very difficult to perform. Assistance could be achieved by using motorized endoscopes and real-time visual tracking algorithm to automatically follow a selected target. In order to control the motors, one needs to have an accurate estimation of the motion of the target in the endoscopic view, which requires an efficient tracking algorithm. In this paper, we compare tracking algorithms on various in vivo targets in order to assess their behavior under different conditions. The study shows that among the difficulties which arise when tracking an in vivo target, the change of illumination is paramount. Nevertheless, some algorithms, with minor modifications and without a priori knowledge about the target, achieve very good results.