What you can’t see can help you – extended-range imaging for 3D-mask presentation attack detection

Abstract

High-quality custom-made 3D masks are increasing becoming a serious threat to face recognition systems. This threat is driven, in part, by the falling cost of manufacturing such masks. Research in face presentation-attack detection (PAD) in general, and also specifically for 3D-mask based attacks, has mostly concentrated on imagery in the visible-light range of wavelengths (RGB). We look beyond imagery in the visible-light spectrum to find potentially easier solutions for the challenge of face presentation-attack detection (PAD). In particular, we explore the use of nearinfrared (NIR) and thermal imagery to detect print-, replay-, and 3D-mask-attacks. This preliminary study shows that both NIR and thermal imagery can potentially simplify the task of face-PAD.