Research map / Adversarial robustness and security / Attacks on text and graphs
Artificial Immune System of Secure Face Recognition Against Adversarial Attacks
Min Ren, Yunlong Wang, Yuhao Zhu, Yongzhen Huang, Zhenan Sun, Qi Lecky Li, Tieniu Tan
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 |
|---|---|---|---|
| add self-supervised adversarial training (ablation) | Ours (AIS defense) · LFW PGD | EER: 40.60% -> 14.19% | Improved |
| add self-supervised adversarial training (ablation) | Ours (AIS defense) · LFW FGSM | EER: 11.99% -> 4.46% | Improved |
| add self-supervised adversarial training (ablation) | Ours (AIS defense) · LFW DeepFool | EER: 12.08% -> 5.01% | Improved |
| add self-supervised adversarial training (ablation) | Ours (AIS defense) · LFW clean | EER: 1.04% -> 1.01% | No clear effect |
| add memory module (ablation) | Ours (AIS defense) · LFW PGD | EER: 22.41% -> 14.19% | Improved |
| add memory module (ablation) | Ours (AIS defense) · LFW FGSM | EER: 9.06% -> 4.46% | Improved |
| add memory module (ablation) | Ours (AIS defense) · LFW DeepFool | EER: 9.75% -> 5.01% | Improved |
| add memory module (ablation) | Ours (AIS defense) · LFW clean | EER: 1.06% -> 1.01% | No clear effect |
| use the proposed defense instead of no defense | ArcFace (ResNet-50) · LFW PGD | EER: 99.71% -> 14.19% | Improved |
| use the proposed defense instead of PIN | ArcFace (ResNet-50) · LFW PGD | EER: 29.77% -> 14.19% | Improved |
| use the proposed defense instead of PIN | ArcFace (ResNet-50) · MegaFace PGD | EER: 33.37% -> 18.63% | Improved |
| use the proposed defense instead of PIN | ArcFace (ResNet-50) · MegaFace DeepFool | EER: 9.20% -> 9.17% | No clear effect |
Closest papers by meaning
- Modeling Biological Immunity to Adversarial Examples (CVPR 2020 · similarity 0.86)
- Efficient Decision-Based Black-Box Adversarial Attacks on Face Recognition (CVPR 2019 · similarity 0.85)
- Phase-aware Adversarial Defense for Improving Adversarial Robustness (ICML 2023 · similarity 0.83)
- An Introduction to Adversarially Robust Deep Learning (TPAMI 2023 · similarity 0.82)
- Adversarial Robust Safeguard for Evading Deep Facial Manipulation (AAAI 2024 · similarity 0.82)
- Non-Adaptive Adversarial Face Generation (NeurIPS 2025 · 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.
