Abstract: In this study, a new model named: Random Pulling Model (RPM) is presented as an alternative feature extraction algorithm for use in automatic face recognition/authentication tasks. We show that the promising RPM algorithm extracts from faces features that are relevant and efficient for authentication. The feasibility of the RPM method has been successfully tested on face authentication using 2360 XM2VTS frontal face images corresponding to 295 subjects, which were acquired under variable illumination and facial expressions.