Visual object tracking using discriminative correlation filter
V. Ramalakshmi, M. Germanus Alex · 2016
Video recording systems are becoming ubiquitous today towards several critical applications including automated video surveillance, traffic monitoring, and smart environments. Visual object tracking is one of challenging and emerging research topics in the recent past in the domain of image processing and computer vision. It is the process of locating and tracking the motion and orientation of one or multiple moving objects over time in the given video sequence. In this paper, we present the design and implementation of a visual object tracking system using video scenes captured from a single surveillance camera. The proposed system uses a discriminative correlation filter based model which is robust and computationally efficient for real time tracking. The proposed system is developed as IPython notebook using several open source libraries. Experiments are conducted on three real world publicly available benchmarking video sequences. Our experimental results show that maximum center location error is only 5 pixels while achieving an average success rate of 80% in tracking the object of the interest.