This table lists the benchmark results for the Binary Classification scenario.


Method InfoDeepfakesFace2FaceFaceSwapNeuralTexturesPristineTotal
sort bysort bysort bysort bysorted bysort by
DirechletEnsemble-Classifier0.9550.9560.9320.9670.9920.973
Top8-soft-withoutnt0.3000.0660.9130.0070.9740.624
Cancer0.9640.7810.9420.7800.9520.903
lcy20.9000.0730.8450.0530.9500.679
Beijing ZKJ1.0000.9200.9610.8800.9480.941
YjzV40.2000.1530.1070.7070.9380.629
ZAntiFakeBio1.0000.9200.9710.9070.9360.940
Leo1.0000.8610.9710.8530.9220.917
MixingExpert1.0000.9050.9420.7800.9220.909
Firefly0.9550.7300.8740.6670.9200.855
Yjz0.9640.6640.9420.5530.9200.837
FAKEDET0.9640.8320.9220.6870.9180.877
Yolo with Voting Mechanism0.8550.6060.7090.5870.9160.796
jiayoua0.3270.8030.1170.7400.9080.723
NoSenseAtAll0.9820.9050.9510.8270.9080.908
gg20.3270.8030.1170.7400.9080.723
RobustForensics0.9910.8910.9510.8070.9040.902
AllDataRes_EB50.9640.8180.8450.6470.9000.852
b7_e2_ensemble_chun0.9270.7590.8060.6870.9000.842
ENbsoftFW11110.9820.8910.9220.7870.8980.892
ATDETECTOR0.9550.7960.9220.7800.8980.875
winer_ensemble_correct20.8730.7450.8160.5600.8960.814
PredictFake0.9730.8470.9130.8200.8940.887
lcy0.9640.8830.8930.5730.8900.850
yangjiezhi-ensemble0.9270.7450.8160.7000.8840.835
Aquarius1.0000.8540.9710.8070.8840.890
StableForensics0.9910.8470.9510.7870.8840.883
effb7_newdataaug_v7_aligne_v20.9450.8100.9420.6130.8840.846
InTeLe_1.0000.7960.9220.7200.8840.864
b7_514_e2_luo_yang_efficienet0.9360.8250.8540.7000.8840.851
en7_yjz0.9730.8980.9220.7200.8800.873
FullEfficientNet0.8360.5260.7280.4800.8800.751
ZFake0.9450.7370.8640.6930.8800.838
yjzenb_lowqulity0.9730.9050.9320.7270.8800.876
696_723_yjz0.9640.8180.8930.7400.8780.860
EnbFW0.9730.9120.9510.7270.8780.878
Balance0.9180.8610.8830.7330.8760.858
CONG0.9550.8100.8540.7800.8740.858
EfficienNetb70.9730.9050.9220.7200.8680.868
result_all_best_th0.85_517_ens0.9450.8470.8540.7400.8660.852
HKU_EfficientNet-B7_v20.8450.6060.8060.4470.8640.758
framelevel_all_compress_format0.7820.1240.1550.1730.8600.575
Yolo Object detection0.8640.6280.7570.6670.8540.786
shangci0.9730.8760.9420.7470.8540.863
KBNet0.9090.7740.8540.7930.8520.839
HRC0.8270.6130.5730.6600.8480.757
iouiwc-310.9640.7960.8930.7600.8480.845
EfficientB5_OHEM0.9180.8760.9130.6530.8480.837
LGSC_Lite0.9550.8390.8740.7730.8440.848
Yjz_jerry0.9730.8760.8930.7130.8400.846
focbe0.9000.8320.7860.7930.8280.827
IAMFAKE20.4640.4960.5150.3670.8260.640
IAMFAKE0.4360.4090.5240.3000.8160.611
unet+res0.8820.8320.7960.7670.8120.814
UltraVision0.9180.8470.9320.7330.8100.828
YJZ_419_513_ensemble0.9640.8320.9510.7800.8080.839
fakeface0.9550.7660.8830.7470.8060.816
effb7_newdataaug_v7_aligne0.9730.8250.9610.6600.8040.820
YJZ_ensemmble0.9360.8100.9420.7530.7940.821
EfficientB4_3800.8820.7010.7960.7400.7940.783
IR-Capsule0.9730.8180.9420.7930.7920.831
RealFace0.9450.7660.8640.8130.7900.815
IR+unet0.9090.8470.8450.7870.7900.816
resnet500.5270.5040.4080.3870.7860.620
Xceptio InceptionRes Efficient0.9640.8250.8640.8270.7680.816
inceptionv10.6550.3500.5440.3330.7680.610
antifake0.9360.8030.9420.6870.7640.795
SCANcopyleft0.9090.8250.7380.6130.7640.763
yd dust face0.9550.8610.9510.6930.7620.806
framelevel0.8730.7370.8350.7600.7600.777
Inception Resnet V1permissive0.9360.8390.9030.8200.7500.809
Nika Dogonadze, Jana Obernosterer: Deep Face Forgery Detection. Advanced Deep Learning for Computer Vision Course at TUM
SiameseDF0.8640.7230.7090.6670.7500.742
DFSpot-Ensemble0.9000.7660.8250.7000.7500.769
Siam-ensemble0.9450.9340.9510.7470.7480.816
simple policybinary0.9550.8690.9420.3870.7440.751
face single model0.9640.8690.9420.4600.7380.760
faceClassify10.9450.8540.9810.7930.7300.806
MesoNet0.8730.5620.6120.4070.7260.660
Darius Afchar, Vincent Nozick, Junichi Yamagishi, and Isao Echizen: Mesonet: a compact facial video forgery detection network. arXiv
swap_classify0.9090.7590.8640.7470.7240.767
EfficientNet-b40.9550.7960.8250.8270.7120.779
EffB7_DataAug_v10.9450.8470.8250.8130.7120.783
EffB7_DataAug_v70.9640.9200.9510.8670.7100.815
AttentionExcep0.7820.7230.5240.7470.7000.701
LVLNet0.9640.7520.7960.6870.6940.741
xceptattention0.7910.7740.4560.7470.6800.692
EfficientNetTest0.7090.4450.6310.3870.6740.599
EffB8_FFDFDCDFCelebAISG0.3820.4090.4470.4330.6600.539
Xception_ForgeryNet0.8180.4960.5440.4070.6320.591
EfficientNetB70.8360.8470.8740.7470.6300.725
PPPP0.8270.6350.6210.4800.6280.628
Sentinel0.9640.9050.8830.8670.6240.763
EfficientNet-b7-Attention0.9550.8030.8540.6000.6160.701
eff-b7-att epc110.9550.7880.8640.5470.5920.680
MMIG-Net0.9090.8250.8160.8870.5660.713
GAEL-Net0.7180.6860.6310.7070.5620.625
Jae-Yong Baek, Yong-Sang Yoo, Seung-Hwan Bae: Generative Adversarial Ensemble Learning for Face Forensics. IEEE Access
Forged face detection : fCNN0.7910.6420.7090.5130.5440.597
Aditi Kohli, Abhinav Gupta: Detecting DeepFake, FaceSwap and Face2Face facial forgeries using frequency CNN. Multimedia Tools and Application, Springer
Xceptionpermissive0.9640.8690.9030.8070.5240.710
Andreas Rössler, Davide Cozzolino, Luisa Verdoliva, Christian Riess, Justus Thies, Matthias Nießner: FaceForensics++: Learning to Detect Manipulated Facial Images. ICCV 2019
XceptionNet Full Imagepermissive0.7450.7590.7090.7330.5100.624
Andreas Rössler, Davide Cozzolino, Luisa Verdoliva, Christian Riess, Justus Thies, Matthias Nießner: FaceForensics++: Learning to Detect Manipulated Facial Images. ICCV 2019
Rahmouni0.8550.6420.5630.6070.5000.581
Nicolas Rahmouni, Vincent Nozick, Junichi Yamagishi, and Isao Echizen: Distinguishing computer graphics from natural images using convolution neural networks. IEEE Workshop on Information Forensics and Security,
DFirt0.9820.9420.9610.8930.4940.717
YSNet0.5640.5840.5920.4270.4920.513
p-DARTS, generalized-cells0.7910.7300.8160.7200.4780.618
Jordi Moreno: Progressive Differentiable Architecture Search for DeepFake Detection. MSc Research Project, from Upf
Bayar and Stamm0.8450.7370.8250.7070.4620.616
Belhassen Bayar and Matthew C. Stamm: A deep learning approach to universal image manipulation detection using a new convolutional layer. ACM Workshop on Information Hiding and Multimedia Security
EfficientNet-v00.8270.7300.8060.5330.4200.564
Recasting0.8550.6790.7380.7800.3440.552
Davide Cozzolino, Giovanni Poggi, and Luisa Verdoliva: Recasting residual-based local descriptors as convolutional neural networks: an application to image forgery detection. ACM Workshop on Information Hiding and Multimedia Security
Steganalysis Features0.7360.7370.6890.6330.3400.518
Jessica Fridrich and Jan Kodovsky: Rich Models for Steganalysis of Digital Images. IEEE Transactions on Information Forensics and Security
eff-b7-v30.9730.9120.9130.8070.1980.546
Two-stream-SRM-RGB10.9820.9270.9420.9530.1740.562
SRTNet_formal_v20.9820.9560.9510.9600.0660.514