An Eye Tracking Scheme Employing Viola-Jones and Template Matching Algorithm

The purpose of this paper is to develop a scheme to track the movement of the eyes. Most eye tracking systems employ Infrared cameras. Considering the application of our work, we are using a regular web camera to minimize the ill effects of IR cameras. Since the implementation is in real time, OpenCV (Open Source Computer Vision) and Code::Blocks are preferred with an added advantage of being free, thus making the system economical. Analysis of the results concluded that while the accuracy of the eye tracker was adequate, more robust system can be developed by using alternate search techniques. This work allows us to develop an alternate mode of communication for those suffering from motor neuron diseases such as ALS (Amyotrophic Lateral Sclerosis). This effort would help PALS (People with ALS) to communicate through their eyes, which is the only mode of interaction available for them.