Automated Face Recognition

Access control and authentication techniques were developed within the framework of Face Recognition. The corresponding face recognition tasks considered herein include, (i) surveilling a gallery of images for the presence of specific probes, and (ii) CBIR subject to correct ID ('match') displaying specific facial landmarks such as wearing glasses. We describe a novel approach for fully automated face recognition and show its feasibility on a large data base of facial images (FERET). Our approach, based on a hybrid architecture consisting of an ensemble of connectionist networks - radial basis functions (RBF) - and inductive decision trees (DT), combines the merits of 'discrete and abstractive' features with those of 'holistic' template matching'. Training for face detection takes place over both positive and negative examples. The benefits of our architecture include (i) detection of faces using decision trees, and (ii) robust face recognition using consensus methods over ensembles of RBF networks. Experimental results, proving the feasibility of our approach, yield (i) 96% accuracy, using cross validation, for surveillance on a data base consisting of 904 images corresponding to 350 subjects, and (ii) 93% accuracy, using cross validation, for CBIR subject to correct TD match tasks on a data base of 200 images.