In this paper, a Bayesian method for face recognition is proposed based on Markov Random Fields (MRF) modeling. Constraints on image features as well as contextual relationships be...
Current two-dimensional image based face recognition systems encounter difficulties with large facial appearance variations due to the pose, illumination and expression changes. ...
In this paper we propose a new approach to compute the scale space of any omnidirectional image acquired with a central catadioptric system. When these cameras are central they ar...
The main problems of Optical Character Recognition (OCR) systems are solved if printed latin text is considered. Since OCR systems are based upon binary images, their results are ...
This paper describes a patch-based approach for rapid image correlation or template matching. By representing a template image with an ensemble of patches, the method is robust wi...