search

VISION-SJTU / RECCE

[CVPR2022] End-to-End Reconstruction-Classification Learning for Face Forgery Detection
MIT License
103 stars 14 forks source link

reported AUC #24

Closed elsobhano closed 11 months ago

elsobhano commented 11 months ago

Hi I have a question about Your results in tabel 4. when you are reporting AUC, does your test datsets contain real images? Becaus in terms of AUC we need to positive and negative samples.

XJay18 commented 11 months ago

Hi, Thanks for your interest.

In Table 4 of the manuscript, we evaluate the model with real images and the specific forged images from the test set. For example, "Train (DF) Test (F2F)" means that we train the model with real images and DF images from the training set and test the model with real images and F2F images from the test set.

XJay18 commented 11 months ago

Close due to inactivation. Please feel free to reopen this issue if you still have related problems.