SLADD

SLADD copied to clipboard

Question about training

22

Anybody have reproduced the training process of the model?

phinoo

Pretrained weights

5

Hi, Can you please share the pretrained weights that you have used to report numbers in the paper? Thanks

joemathai

How long does it take when training？

1

How long does it take when training with NT on the FF++ dataset? This is when using the default parameters of the code.

Takase-Syunki

After so much time, why the weights in the paper are still not released？

5

After so much time, why the weights in the paper are still not released？This puzzled me.

githuboflk

Difficulties reproducing the results

3

Hello, Thank you for your exciting work. I have been trying to reproduce the results of the method, but I was unsuccessful. I have run it on a single GPU...

NamelessGaki

Generation of composite image

5

How to generate composite pictures

zpshs

face detect failed by dlib

16

 how to extract faces from video。i got margined boxes from source video by retinaface model to crop faces and masks that seems be not work。 ``` import os import...

Pudge-tao

In the 45 lines of code of the file src/networks/synthesizer.py, the value of ‘cols’ should be modified instead of the value of ‘rows’, right?

WHUifan

Can you help me to make a mask images for each datasets? I'm confused what code and inputs do I need to enter to create them?

Hyunjoon97

First, thanks for your work. but I want to know how is optimizer_synthesizer work. I can't understand the loss of synthesizer. Can someone explain the following code. At around line...

chenjunlinlin