Research map / Adversarial robustness and security / Attacks on text and graphs
Unified Adversarial Augmentation for Improving Palmprint Recognition
Jianlong Jin, Chenglong Zhao, Ruixin Zhang, Sheng Shang, Yang Zhao, Jun Wang, Jingyun Zhang, Shouhong Ding, Wei Jia, Yunsheng Wu
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 spatial transformation augmentation (ablation) | ArcFace baseline · MPD open-set | TAR@FAR=1e-5: 0.4677 -> 0.6266 | Improved |
| add palmprint generation network augmentation (ablation) | ArcFace baseline · MPD open-set | TAR@FAR=1e-5: 0.4677 -> 0.5514 | Improved |
| add generation network to spatial transformation (ablation) | UAA · MPD open-set | TAR@FAR=1e-5: 0.6266 -> 0.7557 | Improved |
| add spatial transformation to generation network (ablation) | UAA · MPD open-set | TAR@FAR=1e-5: 0.5514 -> 0.7557 | Improved |
| add momentum-based dynamic sampling (ablation) | UAA · MPD open-set | TAR@FAR=1e-5: 0.7557 -> 0.8075 | Improved |
| add momentum-based dynamic sampling (ablation) | UAA · XJTU-UP open-set | TAR@FAR=1e-5: 0.9205 -> 0.9332 | Improved |
| add momentum-based dynamic sampling (ablation) | UAA · BJTU open-set | TAR@FAR=1e-5: 0.7860 -> 0.8192 | Improved |
| add momentum-based dynamic sampling (ablation) | UAA · BJTU open-set | TAR@FAR=1e-6: 0.7052 -> 0.7892 | Improved |
| apply UAA-s instead of ARoFace augmentation | ArcFace · MPD open-set | TAR@FAR=1e-5: 0.6367 -> 0.8075 | Improved |
| apply UAA-s instead of baseline on constrained data | ArcFace · CASIA constrained | TAR@FAR=1e-5: 0.8753 -> 0.9841 | Improved |
| apply RandAugment instead of baseline | ArcFace · PolyU constrained | TAR@FAR=1e-5: 0.9641 -> 0.9603 | Hurt |
| apply AdvProp instead of baseline | ArcFace · MPD open-set | TAR@FAR=1e-6: 0.4000 -> 0.3958 | Hurt |
Closest papers by meaning
- Diff-Palm: Realistic Palmprint Generation with Polynomial Creases and Intra-Class Variation Controllable Diffusion Models (CVPR 2025 · similarity 0.88)
- BézierPalm: A Free Lunch for Palmprint Recognition (ECCV 2022 · similarity 0.88)
- FlowPalm: Optical Flow Driven Non-Rigid Deformation for Geometrically Diverse Palmprint Generation (CVPR 2026 · similarity 0.87)
- RPG-Palm: Realistic Pseudo-data Generation for Palmprint Recognition (ICCV 2023 · similarity 0.86)
- PCE-Palm: Palm Crease Energy Based Two-Stage Realistic Pseudo-Palmprint Generation (AAAI 2024 · similarity 0.84)
- PVTree: Realistic and Controllable Palm Vein Generation for Recognition Tasks (AAAI 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.
