Paper Title
Human Tracking using Template Matching Methodology

Human tracking system has become a very important where security is the priority. Tracking the object in the frames from the video is a key research topic in the computer vision community. The proposed system consist of targeting human face of interest, where we investigate long-term tracking of that face in a video clips. We are using Template Matching Algorithm which is novel tracking framework (TLD) that explicitly decomposes the long-term tracking task into tracking, learning and detection. The tracker follows the object from frame to` frame. The detector localizes all appearances that have been observed so far and corrects the tracker if necessary. A P-N learning method estimates the errors by P-expert which estimates missed detections, and N-expert which estimates false alarms. The learning estimates detector´┐Żs errors and updates it to avoid these errors in the future. Experimental results shows that our approach achieves good performance on video sequences. Keywords - Template, Learning, Detection, Bounding Box, Convolution, diffeomorphism.