Enhancing eID card mobile-based authentication through 3D facial reconstruction

• DOI: 10.1016/j.jisa.2023.103577
• Publication Date: 2023-08-11T11:34:23Z
• Authors: Pietro Ruiu; Andrea Lagorio; Marinella Cadoni; Enrico Grosso
• Keywords: NIST

In recent years the number of credentials and digital identities needed to access online services is exploding, increasing risk fraud unauthorized personal data.Biometric verification techniques can help build more secure authentication systems; in particular face recognition algorithms have reached near perfect accuracy under ideal conditions. Europe 60% population owns an electronic identification (eID) document which contains biometric data, almost all member states provide tools verify identity a user who holds eID card; however, adoption schemes still limited. Reasons for this low are privacy concerns, about storage or transmission images, degradation performance facial with 2D images acquired "in wild" (real world condition inconsistency lighting positioning, unclear occluded features, noise from background). paper new scheme proposed, exploiting 3D reconstruction improving quality probed as changes position limited impact on shape. Several experiments been conducted order investigate proposed method, using Bosphorus NIST Face Recognition Grand Challenge (FRGC) datasets. particular, one most accurate ML has selected compare snapshot obtained back-projection, against uncontrolled wild". We show that reconstructions significantly improve performance, allowing robust system.