Deepfake Detection on FaceForensics++ NT c23
0.9958AUROCPCL + I2G
Evaluation Results
| Method | Links | |
|---|---|---|
| PCL + I2GType of Model=Closed-set, Training Data=FF++c232025.03 | 0.9958 | |
| Face X-rayType of Model=Closed-set, Training Data=FF++c232025.03 | 0.9927 | |
| S-MIL-TType of Model=Closed-set, Training Data=FF++c232025.03 | 0.9885 | |
| MILType of Model=Closed-set, Training Data=FF++c232025.03 | 0.9796 | |
| XN-avgType of Model=Closed-set, Training Data=FF++c232025.03 | 0.9729 | |
| ProposedType of Model=Open-set, Training Data=FF++c232025.03 | 0.9614 | |
| ProposedType of Model=Open-set, Training Data=FF++c23, FS excluded2025.03 | 0.951 | |
| ProposedType of Model=Open-set, Training Data=FF++c23, DF excluded2025.03 | 0.9307 | |
| ProposedType of Model=Open-set, Training Data=FF++c23, F2F excluded2025.03 | 0.93 |