Deepfake Detection on FaceSwap (FS) (test)
99.27AUCQAD-E
Evaluation Results
| Method | Links | |
|---|---|---|
| QAD-ESetting=Quality-agnostic, Compression=Video (raw + c23 + c40)2023.09 | 99.27 | |
| QAD-RSetting=Quality-agnostic, Compression=Video (raw + c23 + c40)2023.09 | 99.01 | |
| Rössler et al.Setting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=published code2023.09 | 98.83 | |
| Fang & LinSetting=Quality-agnostic, Compression=Video (raw + c23 + c40)2023.09 | 98.43 | |
| F3NetSetting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=published code2023.09 | 97.82 | |
| MATSetting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=published code2023.09 | 97.82 | |
| BZNetSetting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=pre-trained weights2023.09 | 97.71 | |
| ADDSetting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=pre-trained weights2023.09 | 95.57 | |
| MesoNetSetting=Quality-agnostic, Compression=Video (raw + c23 + c40)2023.09 | 85.17 | |
| SBIsSetting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=pre-trained weights2023.09 | 80.37 |