Face Recognition at Distance

Face recognition involves matching of an individual or a group of individuals to candidates in databases. The face reflects many aspects of the human character, and a picture represents a mere instance of the face. Requiring a picture to identify humans is asking too much, if not the impossible, given the variations of the environment (illumination), sensors (image quality), and the state of mind of the individual (position and expression). Identification is also difficult due to the causal nature of the process; images of the past are asked to identify individuals now. Given the degradation due to age, sensors, and state of mind of the individuals, face recognition is a function of pose, illumination and expression. These three qualities differ with distance, hence measures of success of face recognition systems at close by distances may not hold with recognition at a distance. A vast literature exists on face recognition. Our group is focusing on the front-end of optical-based face recognition systems at distances beyond 100 meters for cooperating and non-cooperating individuals, in environments that are friendly or otherwise.




Prof. Aly A. Farag
Computer Vision & Image Processing Laboratory
University of Louisville, USA


Montag, 22. November 2010, 16 Uhr


Uni Ulm, N27, Raum 2.033 (Videoübertragung zur Otto-von-Guericke Universität Magdeburg)