TY - GEN
T1 - Keyframe and GAN-Based Data Augmentation for Face Anti-Spoofing
AU - Orfao, Jarred
AU - Van der Haar, Dustin
N1 - Publisher Copyright:
© 2023 by SCITEPRESS-Science and Technology Publications, Lda.
PY - 2023
Y1 - 2023
N2 - As technology improves, criminals, find new ways to gain unauthorised access. Accordingly, face spoofing has become more prevalent in face recognition systems, requiring adequate presentation attack detection. Traditional face anti-spoofing methods used human-engineered features, and due to their limited represen-tation capacity, these features created a gap which deep learning has filled in recent years. However, these deep learning methods still need further improvements, especially in the wild settings. In this work, we use generative models as a data augmentation strategy to improve the face anti-spoofing performance of a vision transformer. Moreover, we propose an unsupervised keyframe selection process to generate better candidate samples for more efficient training. Experiments show that our augmentation approaches improve the baseline performance of the CASIA-FASD and achieve state-of-the-art performance on the Spoof in the Wild database for protocols 2 and 3.
AB - As technology improves, criminals, find new ways to gain unauthorised access. Accordingly, face spoofing has become more prevalent in face recognition systems, requiring adequate presentation attack detection. Traditional face anti-spoofing methods used human-engineered features, and due to their limited represen-tation capacity, these features created a gap which deep learning has filled in recent years. However, these deep learning methods still need further improvements, especially in the wild settings. In this work, we use generative models as a data augmentation strategy to improve the face anti-spoofing performance of a vision transformer. Moreover, we propose an unsupervised keyframe selection process to generate better candidate samples for more efficient training. Experiments show that our augmentation approaches improve the baseline performance of the CASIA-FASD and achieve state-of-the-art performance on the Spoof in the Wild database for protocols 2 and 3.
KW - Face Anti-Spoofing
KW - Generative Data Augmentation
KW - Keyframe Selection
UR - http://www.scopus.com/inward/record.url?scp=85174529738&partnerID=8YFLogxK
U2 - 10.5220/0011648400003411
DO - 10.5220/0011648400003411
M3 - Conference contribution
AN - SCOPUS:85174529738
SN - 9789897586262
T3 - International Conference on Pattern Recognition Applications and Methods
SP - 629
EP - 640
BT - ICPRAM 2023 - Proceedings of the 12th International Conference on Pattern Recognition Applications and Methods, Volume 1
A2 - De Marsico, Maria
A2 - Sanniti di Baja, Gabriella
A2 - Fred, Ana L.N.
PB - Science and Technology Publications, Lda
T2 - 12th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2023
Y2 - 22 February 2023 through 24 February 2023
ER -