Deepfake Detection on NeuralTextures (NT) (test)
94.92AUC (%)QAD-E
Evaluation Results
| Method | Links | |
|---|---|---|
| QAD-ESetting=Quality-agnostic, Compression=Video (raw + c23 + c40)2023.09 | 94.92 | |
| QAD-RSetting=Quality-agnostic, Compression=Video (raw + c23 + c40)2023.09 | 91.25 | |
| Rössler et al.Setting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=published code2023.09 | 89.64 | |
| Fang & LinSetting=Quality-agnostic, Compression=Video (raw + c23 + c40)2023.09 | 89.3 | |
| F3NetSetting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=published code2023.09 | 86.79 | |
| MATSetting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=published code2023.09 | 86.79 | |
| ADDSetting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=pre-trained weights2023.09 | 86.26 | |
| BZNetSetting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=pre-trained weights2023.09 | 80.12 | |
| SBIsSetting=Quality-agnostic, Compression=Video (raw + c23 + c40), Implementation source=pre-trained weights2023.09 | 78.33 | |
| MesoNetSetting=Quality-agnostic, Compression=Video (raw + c23 + c40)2023.09 | 70.24 |