Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Learning Meta Model for Zero- and Few-Shot Face Anti-Spoofing
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use AIM-FAS instead of supervised FAS-DR | FAS-DR · OULU-ZF | ACER (%): 6.60±1.78 -> 4.97±1.29; 4.83±1.40 -> 4.00±1.31; 3.37±1.02 -> 2.44±0.71 | Helped |
| Use AIM-FAS instead of supervised FAS-DR | FAS-DR · Cross-ZF | ACER (%): 13.49±3.62 -> 8.43±2.92; 10.54±3.03 -> 7.34±1.45; 7.21±1.87 -> 3.11±1.01 | Helped |
| Use AIM-FAS instead of supervised FAS-DR | FAS-DR · SURF-ZF | ACER (%): 34.61±2.15 -> 30.97 ±1.28; 33.17±2.07 -> 28.75 ±1.49; 32.50±1.51 -> 27.27±1.25 | Helped |
| Use AIM-FAS with Resnet-10 and binary supervision (ablation) | Resnet-10 · Cross-ZF | ACER (%): 26.51±2.59 -> 14.56±2.63; 26.37±2.64 -> 11.94±1.85; 15.43±2.38 -> 9.02±1.69 | Helped |
| Add predefined living and spoofing faces to the support set (ablation) | AIM-FAS · Cross-ZF 1-shot | ACER (%): 8.25±1.16 -> 7.34±1.45 | Helped |
| Add predefined living and spoofing faces to the support set (ablation) | AIM-FAS · Cross-ZF 5-shot | ACER (%): 7.66±2.45 -> 3.11±1.01 | Helped |
| Use Fusion Training across shot settings (ablation) | AIM-FAS · Cross-ZF 0-shot | ACER (%): 12.61±1.67 -> 8.43±2.92 | Helped |
| Use Fusion Training across shot settings (ablation) | AIM-FAS · Cross-ZF 1-shot | ACER (%): 9.15±1.73 -> 7.34±1.45 | Helped |
| Use Fusion Training across shot settings (ablation) | AIM-FAS · Cross-ZF 5-shot | ACER (%): 3.25±1.05 -> 3.11±1.01 | Hurt |
| Use adaptive inner-update learning rates (ablation) | AIM-FAS · Cross-ZF 0-shot | ACER (%): 11.67±2.08 -> 8.43±2.92 | Helped |
| Use adaptive inner-update learning rates (ablation) | AIM-FAS · Cross-ZF 1-shot | ACER (%): 11.53±2.96 -> 7.34±1.45 | Helped |
| Use adaptive inner-update learning rates (ablation) | AIM-FAS · Cross-ZF 5-shot | ACER (%): 6.44±1.57 -> 3.11±1.01 | Helped |
Closest papers by meaning
- SLIP: Spoof-Aware One-Class Face Anti-Spoofing with Language Image Pretraining (AAAI 2025 · similarity 0.91)
- FaceShield: Explainable Face Anti-Spoofing with Multimodal Large Language Models (AAAI 2026 · similarity 0.90)
- One-Class Face Anti-spoofing via Spoof Cue Map-Guided Feature Learning (CVPR 2024 · similarity 0.89)
- From Intuition to Investigation: A Tool-Augmented Reasoning MLLM Framework for Generalizable Face Anti-Spoofing (CVPR 2026 · similarity 0.88)
- Interpretable Face Anti-Spoofing: Enhancing Generalization with Multimodal Large Language Models (AAAI 2025 · similarity 0.88)
- Test-Time Domain Generalization for Face Anti-Spoofing (CVPR 2024 · similarity 0.87)
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.