Question about total variation regularization and training strategy of appearance generative network

[LeoXing1996]

Thanks for this great work, however, I still have some questions about training strategy and loss term.

The training code of appearance generative network in train.sh is as below: https://github.com/JDAI-CV/Down-to-the-Last-Detail-Virtual-Try-on-with-Detail-Carving/blob/85cc6636450af2653e46e157644641aa9ad36665/train.sh#L8

1. This command uses --joint_parse_loss option instead of --mask_tvloss, and total variation regularization term in Equation 5 is not optimized.
2. This command directly uses --joint_all option, which should be only used after the training of the face refinement network.

[AIprogrammer]

Hi, the tv loss doesn't make so many differences, if you want to use this criterion, just set a small loss weight. For joint training, using joint loss in appearance generation part maybe better, but it's not tested.