Towards pan shot face unlock: Using biometric face information fromdifferent perspectives to unlock mobile devices

Abstract

Purpose – Personal mobile devices currently have access to a significantportion of their user’s private sensitive data and are increasinglyused for processing mobile payments. Consequently, securing accessto these mobile devices is a requirement for securing access to thesensitive data and potentially costly services. The authors proposeand evaluate a first version of a pan shot face unlock method: amobile device unlock mechanism using all information available froma 180° pan shot of the device around the user’s head – utilizingbiometric face information as well as sensor data of built-in sensorsof the device. The paper aims to discuss these issues.Design/methodology/approach – This approach uses grayscale 2D images,on which the authors perform frontal and profile face detection.For face recognition, the authors evaluate different support vectormachines and neural networks. To reproducibly evaluate this pan shotface unlock toolchain, the authors assembled the 2013 Hagenberg stereovision pan shot face database, which the authors describe in detailin this article.Findings – Current results indicate that the approach to face recognitionis sufficient for further usage in this research. However, face detectionis still error prone for the mobile use case, which consequentlydecreases the face recognition performance as well.Originality/value – The contributions of this paper include: introducingpan shot face unlock as an approach to increase security and usabilityduring mobile device authentication; introducing the 2013 Hagenbergstereo vision pan shot face database; evaluating this current panshot face unlock toolchain using the newly created face database.

Publication
International Journal of Pervasive Computing and Communications (IJPCC)