EfficientNetB4 + EfficientNetB4ST + B4Att + B4AttST

Reported on 4 benchmarks across 4 tasks · 1 paper · 4 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology2 results

3D ReconstructiononFaceForensics++
AUC· uses extra data· 2020-04-16
0.9444
best: 0.956 (QAD-E)
SOTA
Video Face Manipulation Detection Through Ensemble of CNNs arXiv:2004.07676
3DonFaceForensics++
AUC· uses extra data· 2020-04-16
0.9444
best: 0.956 (QAD-E)
SOTA
Video Face Manipulation Detection Through Ensemble of CNNs arXiv:2004.07676

Audio1 result

DeepFake DetectiononFaceForensics++
AUC· uses extra data· 2020-04-16
0.9444
best: 0.956 (QAD-E)
SOTA
Video Face Manipulation Detection Through Ensemble of CNNs arXiv:2004.07676

Medical1 result

3D Shape Reconstruction from VideosonFaceForensics++
AUC· uses extra data· 2020-04-16
0.9444
best: 0.956 (QAD-E)
SOTA
Video Face Manipulation Detection Through Ensemble of CNNs arXiv:2004.07676