Deepfake Detection On Fakeavceleb 1
Métriques
AP
ROC AUC
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | AP | ROC AUC |
|---|---|---|
| self-supervised-video-forensics-by-audio | 94.2 | 94.5 |
| exploring-temporal-coherence-for-more-general | 92.3 | 93.1 |
| detecting-deepfakes-without-seeing-any | 96.8 | 97.4 |
| av-lip-sync-leveraging-av-hubert-to-exploit | - | - |
| lip-sync-matters-a-novel-multimodal-forgery | - | - |
| leveraging-real-talking-faces-via-self | 95.3 | 97.1 |
| taming-transformers-for-high-resolution-image | 55.0 | 51.8 |
| leveraging-real-talking-faces-via-self | 73.9 | 59.2 |
| avtenet-audio-visual-transformer-based | - | - |
| faceforensics-learning-to-detect-manipulated | 84.8 | 85.3 |
| lips-don-t-lie-a-generalisable-and-robust | 89.4 | 91.1 |
| lip-sync-matters-a-novel-multimodal-forgery | - | - |
| joint-audio-visual-deepfake-detection | 88.8 | 88.1 |