List your company
ResearchFormerly WizWand

Face Forgery Detection on DFDC (test)

84.72Frame-Level AUCSGF-CDNet
62.443268.226674.0179.7934Mar 25, 2026Apr 10, 2026Apr 27, 2026May 14, 2026May 31, 2026Jun 17, 2026Jul 4, 2026
Updated 3mo ago

Evaluation results

MethodLinks
SGF-CDNetVenue=ICMEW 2026, Input Type=Frame, Training Set (Real)=Yes, Training Set (Fake)=Yes2026.0784.72
CoRITVenues=-, Backbone=CLIP-ViT-L/14, Optimizer=SAM, Backbone Status=Frozen2026.0384.5
WMambaVenue=ACM MM 2025, Input Type=Frame, Training Set (Real)=Yes, Training Set (Fake)=Yes2026.0782.97
FreqDebiasVenues=CVPR’252026.0382.4
DFD-FCGVenue=CVPR 2025, Input Type=Video, Training Set (Real)=Yes, Training Set (Fake)=Yes2026.0781.81
ForAdaVenues=CVPR’25, Backbone=CLIP-ViT-L/14, Optimizer=SAM, Backbone Status=Trainable2026.0381.6
UDDVenue=AAAI 2025, Input Type=Frame, Training Set (Real)=Yes, Training Set (Fake)=Yes2026.0781.2
RAEVenue=ECCV 2024, Input Type=Frame, Training Set (Real)=Yes, Training Set (Fake)=Yes2026.0780.2
KNDVenues=MM’252026.0378.2
SeeABLEVenue=ICCV 2023, Input Type=Frame, Training Set (Real)=Yes, Training Set (Fake)=Yes2026.0775.9
UDDVenues=AAAI’252026.0375.8
FreqBlenderVenue=NeurIPS 2024, Input Type=Frame, Training Set (Real)=Yes, Training Set (Fake)=Yes2026.0774.59
ED [13]Venues=TIP’252026.0374.3
AUNetVenue=CVPR 2023, Input Type=Frame, Training Set (Real)=Yes, Training Set (Fake)=Yes2026.0773.82
LSDAVenues=CVPR’242026.0373.6
UniFDVenues=CVPR’23, Backbone=CLIP-ViT-L/14, Optimizer=SAM, Backbone Status=Frozen2026.0373.6
SBIVenue=CVPR 2022, Input Type=Frame, Training Set (Real)=Yes, Training Set (Fake)=Yes2026.0772.42
ED [55]Venues=AAAI’242026.0372.1
UCFVenues=ICCV’232026.0371.9
RecceVenues=CVPR’222026.0371.3
XceptionVenues=ICCV’192026.0370.8
SPSLVenues=CVPR’212026.0370.4
FFDVenues=CVPR’202026.0370.3
F3NetVenues=AAAI’202026.0370.2
SRMVenues=CVPR’212026.0370
EfficientB4Venues=ICML’192026.0369.6
FoCusVenues=TIFS’242026.0366.9
Face X-rayVenues=CVPR’202026.0363.3