search

jtchen0528 / PCL-I2G

Unofficial Implementation: Learning Self-Consistency for Deepfake Detection
56 stars 7 forks source link

Missing code for generating FF++dataset mask #13

Open CJL0114 opened 8 months ago

CJL0114 commented 8 months ago

I trained a model using the I2G dataset, but the AUC was very low. The paper states that it is necessary to mix the FF++dataset with the I2G dataset for training. The paper states that”The first type is the deepfake video frame from the existing deepfake datasets, for which we find the corresponding real video frame and compute the structural dissimilarity (DSSIM) [60] between them. The mask M is then generated by taking a Gaussian blur of DSSIM followed by thresholding.“ However, it seems that this part of the code is not provided in the paper, resulting in the FF+dataset not having a mask. Can you provide guidance.

jtchen0528 commented 8 months ago

The Gaussian blur may be found here. The way I'm finding the faces (DSSIM) is using a face detector. You may try DSSIM.

Hope this helps.

CJL0114 commented 8 months ago

高斯模糊可以在这里找到。我寻找面部（DSSIM）的方法是使用面部检测器。您可以尝试一下DSSIM。

希望这可以帮助。 Thank you very much！I will continue to complete according to your guidance. If I encounter any problems, I will contact you via email. Thank you for taking the time out of your busy schedule to provide me with careful and helpful guidance.