Research map / Adversarial robustness and security / Backdoor and poisoning
Statistical Meta-Analysis of Presentation Attacks for Secure Multibiometric Systems
Battista Biggio, Giorgio Fumera, Gian Luca Marcialis, Fabio Roli
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| train SVM-RBF on impostor scores resampled with meta-model high-impact attacks (alpha-SVM-RBF) | SVM-RBF · simulated high-impact attacks | GFAR_H1,H2 (%): 25.7 -> 14.0 | Improved |
| add meta-model high-impact attack simulation to LLR (alpha-LLR) | LLR · simulated high-impact attacks | GFAR_H1,H2 (%): 23.4 -> 19.7 | Improved |
| use alpha-LLR instead of Extended LLR (pessimistic p(S_F)=p(S_G)) | Extended LLR · LivDet15+CASIA attacks | GFAR (%): 13.7 -> 8.8 | Improved |
| use alpha-LLR instead of Uniform LLR | Uniform LLR · LivDet15+CASIA attacks | GFAR (%): 13.2 -> 8.8 | Improved |
| use alpha-SVM-RBF instead of SVM-RBF | SVM-RBF · LivDet15+CASIA attacks | GFAR (%): 11.9 -> 9.3 | Improved |
| use alpha-SVM-RBF instead of SVM-RBF | SVM-RBF · LivDet15 fingerprint attack | SFAR (%): 42.0 -> 17.9 | Improved |
| use alpha-SVM-RBF instead of SVM-RBF | SVM-RBF · CASIA face attack | SFAR (%): 5.5 -> 19.1 | Hurt |
| use alpha-LLR instead of LLR | LLR · LivDet15 fingerprint attack | SFAR (%): 37.8 -> 31.9 | Improved |
| use Product rule instead of alpha-SVM-RBF | alpha-SVM-RBF · LivDet15+CASIA attacks | GFAR (%): 9.3 -> 7.1 | Improved |
| use Minimum rule instead of Sum | Sum · no attack | FAR (%): 0.0 -> 3.8 | Hurt |
| use trained secure rules instead of standard rules | standard fusion · no attack | FAR (%): 0.0 -> 0.4 | Hurt |
Closest papers by meaning
- Breaking High Confidence: Practical Face Impersonation under High-Security Thresholds (ECCV 2026 · similarity 0.86)
- SHADOW: Secure Hidden Authenticating Digital Objects in the Wild (TPAMI 2026 · similarity 0.84)
- t-EER: Parameter-Free Tandem Evaluation of Countermeasures and Biometric Comparators (TPAMI 2023 · similarity 0.83)
- Non-Adaptive Adversarial Face Generation (NeurIPS 2025 · similarity 0.81)
- Edit Away and My Face Will not Stay: Personal Biometric Defense against Malicious Generative Editing (CVPR 2025 · similarity 0.81)
- FaceShield: Explainable Face Anti-Spoofing with Multimodal Large Language Models (AAAI 2026 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
