The question of training for BFR

[jiangjin1246]

Hi, thanks for your remarkable work. I've several questions as follows:

As the homepage says, "For face image restoration, we adopt the degradation model used in DifFace for training and directly utilize the SwinIR model released by them as our stage1 model." So, do you only need to train the stage2 model on FFHQ dataset based on the face_swinir_v1.ckpt ?

Thanks again and look forward to your feedback.

[0x3f3f3f3fun]

Yes~