PROJECT TITLE:

Face Spoof Detection With Image Distortion Analysis - 2015

ABSTRACT:

Automatic face recognition is now widely employed in applications starting from deduplication of identity to authentication of mobile payment. This popularity of face recognition has raised concerns about face spoof attacks (also called biometric sensor presentation attacks), where a photo or video of an authorized person's face may be used to gain access to facilities or services. Whereas a range of face spoof detection techniques are proposed, their generalization ability has not been adequately addressed. We tend to propose an efficient and rather robust face spoof detection algorithm primarily based on image distortion analysis (IDA). Four completely different options (specular reflection, blurriness, chromatic moment, and color diversity) are extracted to create the IDA feature vector. An ensemble classifier, consisting of multiple SVM classifiers trained for different face spoof attacks (e.g., printed photo and replayed video), is used to distinguish between real (live) and spoof faces. The proposed approach is extended to multiframe face spoof detection in videos using a voting-based mostly scheme. We have a tendency to additionally collect a face spoof database, MSU mobile face spoofing database (MSU MFSD), using 2 mobile devices (Google Nexus 5 and MacBook Air) with 3 sorts of spoof attacks (printed photo, replayed video with iPhone 5S, and replayed video with iPad Air). Experimental results on two public-domain face spoof databases (Idiap REPLAY-ATTACK and CASIA FASD), and therefore the MSU MFSD database show that the proposed approach outperforms the state-of-the-art methods in spoof detection. Our results also highlight the difficulty in separating genuine and spoof faces, particularly in cross-database and cross-device eventualities.