Research map / Label-efficient and robust learning / Domain adaptation and generalization
Rethinking Domain Generalization for Face Anti-Spoofing: Separability and Alignment
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 separability and alignment together in SA-FAS | Cross-domain face anti-spoofing · OCM→I | HTER: 8.88 -> 6.58 | Helped |
| Use separability and alignment together in SA-FAS | Cross-domain face anti-spoofing · ICM→O | HTER: 11.82 -> 10.00 | Helped |
| Use separability and alignment together in SA-FAS | Cross-domain face anti-spoofing · OCI→M | HTER: 6.67 -> 5.95 | Helped |
| Use separability and alignment together in SA-FAS | Cross-domain face anti-spoofing · OMI→C | HTER: 10.00 -> 8.78 | Helped |
| Add PG-IRM alignment to SupCon separability (ablation) | Cross-domain face anti-spoofing · Four cross-domain settings, average | HTER: 17.03 -> 14.25 | Helped |
| Add SupCon separability to PG-IRM (ablation) | Cross-domain face anti-spoofing · Four cross-domain settings, average | HTER: 15.58 -> 14.25 | Helped |
| Add alignment and separability losses to ERM (ablation) | Cross-domain face anti-spoofing · Four cross-domain settings, average | HTER: 17.22 -> 14.25 | Helped |
| Replace SupCon with SimCLR (ablation) | Cross-domain face anti-spoofing with ERM classification · Four cross-domain settings, average | HTER: 17.03 -> 22.53 | Hurt |
| Replace SupCon with SimSiam (ablation) | Cross-domain face anti-spoofing with ERM classification · Four cross-domain settings, average | HTER: 17.03 -> 18.89 | Hurt |
| Replace SupCon with Triplet loss (ablation) | Cross-domain face anti-spoofing with ERM classification · Four cross-domain settings, average | HTER: 17.03 -> 18.75 | Hurt |
| Replace PG-IRM with VREx (ablation) | Cross-domain face anti-spoofing with SupCon · Four cross-domain settings, average | HTER: 14.25 -> 25.02 | Hurt |
| Replace PG-IRM with IRM-v1 (ablation) | Cross-domain face anti-spoofing with SupCon · Four cross-domain settings, average | HTER: 14.25 -> 17.41 | Hurt |
Closest papers by meaning
- Gradient Alignment for Cross-Domain Face Anti-Spoofing (CVPR 2024 · similarity 0.94)
- Test-Time Domain Generalization for Face Anti-Spoofing (CVPR 2024 · similarity 0.93)
- Rethinking Generalizable Face Anti-spoofing via Hierarchical Prototype-guided Distribution Refinement in Hyperbolic Space (CVPR 2024 · similarity 0.91)
- Domain-Hallucinated Updating for Multi-Domain Face Anti-spoofing (AAAI 2024 · similarity 0.90)
- CFPL-FAS: Class Free Prompt Learning for Generalizable Face Anti-spoofing (CVPR 2024 · similarity 0.89)
- SLIP: Spoof-Aware One-Class Face Anti-Spoofing with Language Image Pretraining (AAAI 2025 · similarity 0.89)
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.