Title
Self-supervised 2D face presentation attack detection via temporal sequence sampling
Abstract
•Inter-frame 2D affine motion compensation is exploited for detecting 2D face artefacts.•Temporal Sequence Sampling (TSS) is proposed to encode a video into a single image.•A self-supervised learning scheme is presented for face presentation attack detection.•Promising generalization is achieved in cross-database tests on public benchmarks.
Year
DOI
Venue
2022
10.1016/j.patrec.2022.03.001
Pattern Recognition Letters
Keywords
DocType
Volume
Face recognition,Presentation attack detection,Spoofing,Liveness detection,Self-supervised learning,Motion compensation
Journal
156
ISSN
Citations 
PageRank 
0167-8655
0
0.34
References 
Authors
2
3
Name
Order
Citations
PageRank
Muhammad Usman111.40
Zitong Yu26112.74
Jukka Komulainen300.34