rshaojimmy / seqdeepfake Goto Github PK

I am sorry to disturb you. I am very interested in your paper. Can you provide some code for recovery. Thank you very much

It seems that the downloading link to pre-trained models have been expired. May the authors check it out?

May I ask for a demo code of manipulating facial componnets with StyleMapGAN on the CelebAMask-HQ dataset? Thank you very much.

Recently, I have been studying this paper and I am not very clear about formula 4 in the paper. The spatial weight map M is generated for each type of manipulation operation. Therefore, if there are several annotations in the manipulation sequence, there should be several M. So, in formula 4, M should be the sum of all M for each operation of each input? I am not very clear about this calculation.

Hi, the work you proposed is very interesting. Could you kindly provide me with the code related to the recovery of DeepFakes?Thank you very much.

Hello, the work you proposed is very interesting. I used your open code for training, but the experimental results cannot reach the results in the paper, especially in components on r50, which is 10 points lower than that in the paper. May I ask why?

Today I cannot access to dataset link. It showed that the data link is removed. Can you check it again for me

Thanks for this creative work! I wondering if you can supplement all original faces wiithout forgery in facial attribute dataset as I do not find the corresponding original face in the ffhq dataset.