Face Image Quality Assessment in Electronic ID Documents

Face image quality estimation is still an open issue since, unlike what happens for other biometric characteristics such as fingerprints, no standard definitions are available yet. The problem even harder when the focus of assessment context electronic ID documents which, according to provisions ISO/IEC 39794-5, a value will be stored in future dedicated blocks. In case high images, general indicators literature tend assign flat score that does not contribute provide significant information. This work study aimed at defining indicator high-quality able predict utility specific face verification purposes. A regressor proposed, based on large set elements including ISO/ICAO controls and scores provided by deep-learning solutions. number experiments highlight issued addressed this scenario confirm effectiveness proposed approach with different recognition systems.