Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

SID-Set

Benchmarks

Task NameDataset NameSOTA ResultTrend
Deepfake DetectionSID-Set (val)
Acc (Real)89.6
18
Forgery LocalizationSID-Set Tampered
AUC87.3
5
Local Editing DetectionSID-Set Tampered Fake Images
Accuracy93.7
3
Local Editing DetectionSID-Set Tampered Real Images
Accuracy99.4
3
Global Synthesis DetectionSID-Set FullSync Fake Images
Accuracy99.5
3
Global Synthesis DetectionSID-Set FullSync Real Images
Accuracy99.4
3
Image InterpretabilitySID-Set description (val)
Mask Score2.28
2
Showing 7 of 7 rows