Eye feature point tracking by using graphical models

Official URL: http://dx.doi.org/10.1109/SIU.2007.4298738

In this paper, a statistical method for eye feature point tracking is proposed. The aim is to track feature points even when the observed data are uncertain because of noise and/or occlusion. With this motivation, a graphical model that uses the spatial information as well as the temporal information between points is built. The proposed method is applied on 2D grayscale real video sequences as a real data application.