This paper deals with robust feature tracking.
The aim is to track point features in a sequence of images and to
identify unreliable features. We extend the well-known
Shi-Tomasi-Kanade tracker \cite{[ST94]} by introducing an automatic
scheme for rejecting spurious features in the first image.
We employ a simple and efficient rejection rule based on gray levels
co-occurence entropy and show that its empericaly assumptions are
satisfied in the feature tracking scenario. Experiments with real
and synthetic images confirm that our algorithm makes good features
tracking. We illustrate quantitatively the benefits introduced by
the developped algorithm.
