Research map / Label-efficient and robust learning / Recognition and long-tail
CelebA-Spoof: Large-Scale Face Anti-Spoofing Dataset with Rich Annotations
Yuanhan Zhang, ZhenFei Yin, Yidong Li, Guojun Yin, Junjie Yan, Jing Shao, Ziwei Liu
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 semantic attribute supervision to live/spoof classification (ablation) | baseline ResNet-18 · CelebA-Spoof | ACER (%): 3.8 -> 2.85 | Improved |
| use semantic attributes alone instead of live/spoof label (ablation) | baseline · CelebA-Spoof | ACER (%): 3.8 -> 4.17 | Hurt |
| add spoof type attribute supervision (ablation) | AENetC,S · CelebA-Spoof | APCER (%): 5.31 -> 4.62 | Improved |
| add face attribute supervision (ablation) | AENetC,S · CelebA-Spoof | BPCER (%): 1.37 -> 1.09 | Improved |
| add illumination supervision (ablation) | AENetC,S · CelebA-Spoof | ACER (%): 2.99 -> 2.85 | Improved |
| add depth map supervision as auxiliary (ablation) | baseline · CelebA-Spoof | ACER (%): 3.8 -> 3.26 | Improved |
| use depth map alone for classification (ablation) | baseline · CelebA-Spoof | ACER (%): 3.8 -> 4.51 | Hurt |
| add depth map to AENetC,G (ablation) | AENetC,G · CelebA-Spoof | ACER (%): 3.55 -> 3.26 | Improved |
| add reflection map to AENetC,G (ablation) | AENetC,G · CelebA-Spoof | ACER (%): 3.59 -> 3.26 | Improved |
| add both semantic and geometric supervision | AENetC,S,G · CelebA-Spoof cross-domain Protocol 1 | ACER (%): 4.11 -> 3.09 | Improved |
| add geometric supervision to baseline | AENetC,G · Protocol 2 sensor | ACER (%): 5.05 -> 5.31 | Hurt |
| train on CelebA-Spoof instead of SiW | AENetC,S,G · CASIA-MFSD | HTER (%): 27.6 -> 11.9 | Improved |
Closest papers by meaning
- FaceShield: Explainable Face Anti-Spoofing with Multimodal Large Language Models (AAAI 2026 · similarity 0.89)
- One-Class Face Anti-spoofing via Spoof Cue Map-Guided Feature Learning (CVPR 2024 · similarity 0.89)
- SLIP: Spoof-Aware One-Class Face Anti-Spoofing with Language Image Pretraining (AAAI 2025 · similarity 0.88)
- Interpretable Face Anti-Spoofing: Enhancing Generalization with Multimodal Large Language Models (AAAI 2025 · similarity 0.88)
- mmFAS: Multimodal Face Anti-Spoofing Using Multi-Level Alignment and Switch-Attention Fusion (AAAI 2025 · similarity 0.87)
- From Intuition to Investigation: A Tool-Augmented Reasoning MLLM Framework for Generalizable Face Anti-Spoofing (CVPR 2026 · 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.
