The Third Competition on Document Forgery Detection on ID-Cards and Passports

2026-07-17 · Source: Computer Vision and Pattern Recognition · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cybersecurity & Data Privacy, Computer Vision · Depth: Expert, quick

Summary

The Third International Competition on Document Forgery Detection on ID-Cards and Passports evaluated over 100 submission models from 63 registered teams across two distinct tracks. Track 1 focused on synthetic-data-based ID-PAD systems under controlled yet diverse conditions, where the Incode team achieved an AV_Rank of 27.82%, demonstrating consistent performance and the value of generalizable design. Track 2 presented a more complex challenge with heterogeneous attack scenarios across various domains, and Incode again secured the top position with an AV_Rank of 68.71%. These results underscore that effective Presentation Attack Detection (PAD) requires both high accuracy and robust consistency against diverse attack types and imaging conditions. The competition has become a leading benchmark for secure identity verification.

Key takeaway

For Computer Vision Engineers developing ID document forgery detection systems, prioritize solutions that demonstrate consistent performance across diverse attack types and imaging conditions, not just high accuracy on limited datasets. Your evaluation metrics should emphasize generalizability and robustness, as exemplified by the top-performing systems in this competition. This approach will ensure your PAD systems are truly effective in real-world, heterogeneous scenarios.

Key insights

Effective Presentation Attack Detection demands high accuracy and consistent performance across varied attack types and imaging conditions.

Principles

• PAD effectiveness requires high accuracy and consistency.
• Generalizable design is crucial for consistent performance.
• Collaboration between companies and academia drives progress.

In practice

• Evaluate PAD systems against diverse attack types.
• Prioritize generalizable designs for ID-PAD systems.

Topics

• Document Forgery Detection
• ID-Cards
• Passports
• Presentation Attack Detection
• Computer Vision
• Biometric Security

Best for: Research Scientist, CTO, VP of Engineering/Data, AI Scientist, Computer Vision Engineer, AI Security Engineer

Related on AIssential

• From Vision to Text: A Compact Multimodal Approach for Robust, Cross-Domain Presentation Attack Detection on ID Cards — Compact multimodal PAD models need real-world data and fine-tuning for robust cross-domain performance, challenging synthetic dataset utility.
• Format-Controlled Multi-Scale JPEG Compression Response Analysis for Image-Level Forgery Screening — Lightweight CPU-based feature engineering with multi-scale ELA effectively screens image forgeries by detecting compression-history inconsistencies.
• Temporal Modeling of Optically Variable Devices in Identity Documents — Modeling temporal dynamics of OVDs with self-supervised methods defeats advanced identity document fraud without attack samples.
• The AI Illusion (Part 2): The AI Detection Mirage — Current AI image detectors are unreliable, often misidentifying human art or hallucinating evidence.
• Rethinking Infrastructure Inspection as Image Difference Classification: A Traffic Sign Case Study — Image difference classification (IDC) effectively reduces data reliance for infrastructure inspection in digital twin systems.
• Part 2: Why Traditional CNNs Aren’t Enough for Modern Deepfake Detection — Effective deepfake detection demands a multi-domain forensic approach, moving beyond visible pixels and benchmark-centric evaluation to generalize against evolving AI generators.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Computer Vision and Pattern Recognition.