Visual tracking of objects is one the several capabilities that have the human beings. Even though it is performing in an unconscious way, it is tightly linked with many of the tasks we do.
At the present time, introducing these capabilities in artificial visual systems is one of the most interesting areas to research in computer vision and robotics.
Our effort is focused in the development of algorithms and techniques that allows an automatic adaptation of visual systems to changing environments.
With the purpose of making a complete vision-based control system, it is necessary to integrate several research areas such as visual matching, visual tracking and visual servoing.
The first step towards this integration consists of the use of model vision-based control techniques. These techniques have been designed to control a robot with respect to objects of any shape. Indeed, the system proposed have been designed for positioning a robot by tracking specific visual features (e.g. interest points, straight lines, contours).