Compact Vision Models for Iris Presentation Attack Detection under Presentation Attack Instrument Shift and Environmental Degradation

arXiv:2609.20386 · cs.CV, cs.CR · Submitted 2026-09-17 · Read on arXiv

cs.CV, cs.CR

Submitted: 2026-09-17

Updated: 2026-09-17

Comments: Accepted at BIOSIG 2026. This preprint includes minor nomenclature and editorial corrections clarifying the project-specific Patch-ABMIL and Compact-TransMIL variants

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Benchmark — 8× in this paper · explained in 2 episodes
• Transformer — 7× in this paper · explained in 2 episodes
• Ablation — 5× in this paper · explained in 1 episode
• Adam — 5× in this paper · explained in 2 episodes
• Domain Shift — 4× in this paper · explained in 4 episodes
• Generalization — 4× in this paper · explained in 13 episodes
• Attack Presentation Classification Error Rate — 3× in this paper
• Data Augmentation — 3× in this paper · explained in 7 episodes
• Robustness — 3× in this paper · explained in 5 episodes
• Computer Vision and Pattern Recognition Workshops — 2× in this paper
• Embeddings — 2× in this paper · explained in 2 episodes
• Foundation Model — 2× in this paper · explained in 3 episodes
• Foundation Models — 2× in this paper · explained in 7 episodes
• Binarized Statistical Image Features — 1× in this paper
• Bona Fide Presentation Classification Error Rate — 1× in this paper
• Detection Error Trade-off — 1× in this paper
• Edge Deployment — 1× in this paper · explained in 2 episodes
• IEEE International Joint Conference on Biometrics — 1× in this paper
• Joint Photographic Experts Group — 1× in this paper

Sources

• ZACH-ViT: A Zero-Token Vision Transformer with ShuffleStrides Data Augmentation for Robust Lung Ultrasound Classification
• ZACH-ViT: Regime-Dependent Inductive Bias in Compact Vision Transformers for Medical Imaging
• Iris Liveness Detection Competition (LivDet-Iris) -- The 2023 Edition
• Towards Iris Presentation Attack Detection with Foundation Models
• Attention-based Deep Multiple Instance Learning
• TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classification

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models