VISER: Visually-Informed System for Enhanced Robustness in Open-Set Iris Presentation Attack Detection

2026-03-18

AI Summary

This work addresses the performance degradation in open-set iris presentation attack detection (PAD) caused by unknown attack types by proposing a novel training paradigm that integrates human visual priors to guide model learning. Through a systematic comparison of various saliency cues—including eye-tracking heatmaps, manual annotations, segmentation masks, and DINOv2 embeddings—the study finds that denoised eye-tracking heatmaps most effectively enhance model generalization. Evaluated under the leave-one-attack-type-out protocol, the proposed method significantly outperforms the standard cross-entropy baseline in both AUROC and APCER@BPCER=1% metrics, offering an effective and robust solution for open-set iris liveness detection.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Adversarial Search

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements

Abstract

Human perceptual priors have shown promise in saliency-guided deep learning training, particularly in the domain of iris presentation attack detection (PAD). Common saliency approaches include hand annotations obtained via mouse clicks and eye gaze heatmaps derived from eye tracking data. However, the most effective form of human saliency for open-set iris PAD remains underexplored. In this paper, we conduct a series of experiments comparing hand annotations, eye tracking heatmaps, segmentation masks, and DINOv2 embeddings to a state-of-the-art deep learning-based baseline on the task of open-set iris PAD. Results for open-set PAD in a leave-one-attack-type out paradigm indicate that denoised eye tracking heatmaps show the best generalization improvement over cross entropy in terms of Area Under the ROC curve (AUROC) and Attack Presentation Classification Error Rate (APCER) at Bona Fide Presentation Classification Error Rate (BPCER) of 1%. Along with this paper, we offer trained models, code, and saliency maps for reproducibility and to facilitate follow-up research efforts.

Problem

• iris presentation attack detection
• open-set
• human saliency
• generalization
• robustness

Innovation

• visually-informed learning
• eye tracking heatmaps
• open-set iris PAD
• saliency-guided training
• DINOv2 embeddings