Title :
Face authentication for an activity-related biometrics system
Author :
Salah, Albert Ali
Author_Institution :
CWI - Signals & Images, Amsterdam, Netherlands
Abstract :
This paper presents a preliminary study in the use of face recognition for unobtrusive activity-related biometrics systems. First the components of such a system are described. Several issues related to using the face modality are identified, including the need for an adaptive face localization scheme for robust authentication. A greedy algorithm is proposed to complement the Viola-Jones algorithm for this purpose. Experimental results are presented on a database with 11 subjects (10 sessions per subject) performing various activities.
Keywords :
biometrics (access control); face recognition; greedy algorithms; Viola-Jones algorithm; activity-related biometrics system; adaptive face localization scheme; face authentication; face modality; face recognition; greedy algorithm; Annealing; Authentication; Biometrics; Databases; Face recognition; Greedy algorithms; Iris; Particle filters; Reactive power; Robustness;
Conference_Titel :
Signal Processing and Communications Applications Conference, 2009. SIU 2009. IEEE 17th
Conference_Location :
Antalya
Print_ISBN :
978-1-4244-4435-9
Electronic_ISBN :
978-1-4244-4436-6
DOI :
10.1109/SIU.2009.5136495
