ForgeryNet

ForgeryNet copied to clipboard

Hi, author Could you provide corresponding facial landmarks or bounding box files? That will be very helpful for availability of ForgeryNet

niefengxxx

Data Preprocessing

1

Hello, thank you very much for your outstanding work. I would like to use the ID-remained subset of the dataset, but the data processing code I wrote myself is not...

shimmer-ghq

Exact dataset labels

13

Thank you for releasing the dataset, it is really valuable for deepfake researches. Could you release the labels of testset and the exact manipulation methods of 16cls_labels?

yoctta

About label list

1

Hello, thanks for sharing us such excellent work! Here i have some issues about the label list: taking a video forgery as example, "3 val_video_release/4/7f89846f0181198f2aee02c863c2aca5/718e0e03ab494625ccd35e839deafadb/video.mp4 0 10 0 10 35...

Mark-Dou

Will the labels for test set be released?

3

Currently, it seems that only the images/videos are available on your webpage. I wonder will the labels for the test set be released?

XJay18

Hello, I see that many images have severe perturbations. In the paper you mention that you randomly apply perturbations to 99% of the data. Are there annotations of which perturbations...

raquelroman

validation dataset

1

Hello, the quark disk in the verification set is suspected of violating regulations. Can you send a new one.

zxq99799

Lable of Baseline model

3

Thanks for sharing this gread work! I have downloaded xception baseline model and run it. The model returned a two-class score, I am wondering which one is for fake label?

zoezhu

Error when run main.py

1

hi ,when I git clone your code and run main.py,there is one mistake File "/eve/code/ForgeryNet/intra_dataset_code/xception.py", line 94, in Xception bn_var_mode=syncbnVarMode_t.L2, use_sync_bn=True, bypass_last_bn=False, dropout=0, **kwargs): NameError: name 'syncbnVarMode_t' is not defined...

Eve-Wang

Access Permission of Google drive

1

Hi, Sincerely thanks for your brilliant work! I want to download the training set for research. Would you please give me an access? My email address is [email protected].

lrxsxdl