Algorithm 1: Proposed Tracking Method

Input: Video with initialized ground truth on frame 1.   Output: Rectangle on each frame. for 1st to the last frame.   Compute context prior model by using (3).   Compute confidence map by using (11).   Compute center of target location.   Estimate scale by using (15).   Compute APCE by using (16).   Determine occlusion detection using (17) and (18).   Check four rules of occlusion detection given in Section 3.2.   if rule 2 occurs      Activate fractional-gain Kalman filter   Compute fractional Kalman gain by using (30).   Predict position by using (22).   Compute error covariance by using (28).   end   Calculate occlusion indicator using (31).   Calculate learning rate using (32)   Update context prior model by using (3).   Update spatial context model by using (9).   Update STC model by using (12).   Estimate the position of target. End