List your company
ResearchFormerly WizWand

Deepfake Detection on FF++ (test)

99.96AUCLAA-Net
45.193659.411873.6387.8482Mar 23, 2022Nov 12, 2022Jul 5, 2023Feb 24, 2024Oct 16, 2024Jun 7, 2025Jan 28, 2026
Updated 7mo ago

Evaluation results

MethodLinks
LAA-NetEvaluation Protocol=In-distribution2025.1199.96——
MADDTraining Set=FF++2022.0399.8——
ID-unaware Deepfake Detection ModelBackbone=Efficient-b42022.1099.79——
ID-unaware Deepfake Detection ModelBackbone=Efficient-b32022.1099.78——
XceptionTraining Set=FF++2022.0399.7——
ID-unaware Deepfake Detection ModelBackbone=ResNet-342022.1099.7——
SBIBackbone=Efficient-b42022.1099.64——
MATBackbone=Efficient-b42022.1099.61——
XceptionBackbone=Xception2022.1099.58——
MMMSBackbone=Transformer2022.1099.5——
FauxNetEvaluation Protocol=In-distribution2025.1199.3896—
MARE2026.0199.2897.5—
M2F2Venue=CVPR20252026.0199.2597.38—
Face-x-rayBackbone=HRNet2022.1099.17——
PCL+I2GBackbone=ResNet-342022.1099.11——
SLADDTraining Set=FF++2022.0398.4——
SLADDBackbone=Xception2022.1098.4——
ExpDVenue=AAAI20242026.0198.495.6—
VideoMAEEvaluation Protocol=In-distribution2025.1198.394.57—
F3-NetBackbone=Xception2022.1098.1——
SPSLBackbone=Xception2022.1096.91——
SPSLTraining Set=FF++2022.0396.9——
SMILTraining Set=FF++2022.0396.8——
CapsuleTraining Set=FF++2022.0396.6——
EfficientNetVenue=ICML20192026.0194.1891.19—
EfficientNet B4Evaluation Protocol=In-distribution2025.1193.82——
Two-branchTraining Set=FF++2022.0393.2——
Two-branchBackbone=DenseNet2022.1093.2——
DSP-FWATraining Set=FF++2022.0393——
DSP-FWABackbone=ResNet-502022.1093——
FauxNetEvaluation Protocol=Zero-shot, Training Set=Authentica-Vox2025.1191.4384.43—
SFICVenue=TIFS20242026.0190.390.24—
Meso4Training Set=FF++2022.0384.7——
MesoInception4Training Set=FF++2022.0383——
ID-RevealModality=V, w/ Ref.=✓, Train set=VoxCeleb22025.1081.28—88.19
FWATraining Set=FF++2022.0380.1——
RefereeModality=AV, w/ Ref.=✓, Train set=FakeAVCeleb2025.1079.78—91
Multi-task2022.1076.3——
Two-streamTraining Set=FF++2022.0370.1——
AVADModality=AV, w/ Ref.=✗, Train set=LRS2/LRS32025.1067.01—88.7
VA-MLPTraining Set=FF++2022.0366.4——
XceptionModality=V, w/ Ref.=✗, Train set=FakeAVCeleb2025.1059.48—76.6
POI-ForensicsModality=AV, w/ Ref.=✓, Train set=VoxCeleb22025.1051.33—84.75
HeadposeTraining Set=FF++2022.0347.3——
CNN1# Params=537,9942026.02—82.44—
CNN2# Params=108,6182026.02—79.85—
Mapping CNN (CNN1)# Params=1024, Training Strategy=Full Network training, Base Architecture=CNN12026.02—81.11—
Mapping CNN (CNN1)# Params=2048, Training Strategy=Full Network training, Base Architecture=CNN12026.02—85.23—
Mapping CNN (CNN1)# Params=1956, Training Strategy=Layer-wise training, Base Architecture=CNN12026.02—86.23—
Mapping CNN (CNN1)# Params=2792, Training Strategy=Layer-wise training, Base Architecture=CNN12026.02—88.05—
Mapping CNN (CNN2)# Params=1024, Training Strategy=Full Network training, Base Architecture=CNN22026.02—82.78—
Mapping CNN (CNN2)# Params=2048, Training Strategy=Full Network training, Base Architecture=CNN22026.02—84.09—
Mapping CNN (CNN2)# Params=1872, Training Strategy=Layer-wise training, Base Architecture=CNN22026.02—83.1—
Mapping CNN (CNN2)# Params=2688, Training Strategy=Layer-wise training, Base Architecture=CNN22026.02—86.28—
Mapping Networks# Params=2048, Layers=All2026.02—91.02—
Mapping Networks# Params=1024, Layers=L-4, FC2026.02—89.23—
ResNet50# Params=25M, Layers=All2026.02—91.78—
ResNet50# Params=17M, Layers=L-4, FC2026.02—88.03—