Pose-corrected face processing on video sequences for webcam-based remote biometric authentication

In this paper we describe a framework aimed to perform face-based biometric user authentication for web-resources through client-server secured sessions. A novel front-end for face video sequences processing is developed in which face detection and shot selection is performed at client-side while statistical multi-shot pose-corrected face verification is performed at server-side. We explain all the image processing steps, from the acquisition to the decision, paying special attention to a PDM-based pose correction subsystem and a GMM-based sequence decision test. The pose correction relies on projecting a face-shape mesh onto the set of PDM eigenvectors and back-projecting it after changing the coefficients associated to pose variation. The aligned and discriminatively selected texture features form the observation vectors ready to be pluged into a GMM based likelihood ratio for statistical decision. Tests over known databases show the reliability of The authors are with the Departamento de Teoŕıa de la Señal y Comunicaciones, ETSI Telecomunicación, Universidad de Vigo, 36310 Vigo, Spain. Phone: +34 986 812664. Fax: +34 986 812116. E-mails: {jalba,danisub,eargones,eli,eotero,carmen}@gts.tsc.uvigo.es. Corresponding author. E-mail: [email protected] This work was supported with funds provided partially by the Spanish ministry of education under project TEC2005-07212/TCM and the European sixth framework programme under the Network of Excellence BIOSECURE (IST-2002-507604)