Second Competition on Presentation Attack Detection on ID Card

2025-07-27

AI Summary

This study addresses presentation attack detection (PAD) for identity cards, proposing the first standardized evaluation framework specifically designed for ID-card PAD. Methodologically, it introduces a dual-track evaluation protocol: Track 1 assesses algorithm generalization on a unified benchmark dataset, while Track 2 evaluates dataset discriminability in distinguishing algorithm performance. A new, publicly released ID-card PAD dataset and an automated evaluation platform accompany the framework. Following biometric liveness detection conventions, evaluation employs two primary metrics: Average Rank (AV-Rank) and Equal Error Rate (EER). Results show significant improvements over the inaugural competition: Track 1 achieves 40.48% AV-Rank and 11.44% EER; Track 2 attains 14.76% AV-Rank and 6.36% EER. These contributions advance reproducible, comparable, and rigorous evaluation methodologies for ID-card security verification.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Machine Learning: Evaluation and Analysis
• Search and Optimization: Evaluation and Analysis

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success

Abstract

This work summarises and reports the results of the second Presentation Attack Detection competition on ID cards. This new version includes new elements compared to the previous one. (1) An automatic evaluation platform was enabled for automatic benchmarking; (2) Two tracks were proposed in order to evaluate algorithms and datasets, respectively; and (3) A new ID card dataset was shared with Track 1 teams to serve as the baseline dataset for the training and optimisation. The Hochschule Darmstadt, Fraunhofer-IGD, and Facephi company jointly organised this challenge. 20 teams were registered, and 74 submitted models were evaluated. For Track 1, the "Dragons" team reached first place with an Average Ranking and Equal Error rate (EER) of AV-Rank of 40.48% and 11.44% EER, respectively. For the more challenging approach in Track 2, the "Incode" team reached the best results with an AV-Rank of 14.76% and 6.36% EER, improving on the results of the first edition of 74.30% and 21.87% EER, respectively. These results suggest that PAD on ID cards is improving, but it is still a challenging problem related to the number of images, especially of bona fide images.

Problem

• Evaluating algorithms for ID card presentation attack detection
• Comparing performance of datasets in detecting ID card fraud
• Assessing improvements in ID card security against spoofing

Innovation

• Automatic evaluation platform for benchmarking
• Two tracks for algorithm and dataset evaluation
• New ID card dataset shared for baseline training