vita-group / ssharmonization Goto Github PK

Hi, I encounter "opt" not defined problem.
I can't find where opt is declared.
And the class OPT is empty.
Does that mean user have to define their own OPT ?

Can you please update the license of this repository?

Hi, I encounter the problem that "transform" not is defined.
comp = transform(comp) mask = tf.to_tensor(mask) real = transform(real)

expect

Hi, I've tried to run demo but it doesn't work.
It seems there is no requirements.txt in this repository, could you please add?
Also there is no colour_checker_detection directory in this folder.

Thanks!

Hello.
In the supplementary material, I would like to ask why the input channel of the content network is 4 , is it adding mask in addition to the picture input of three channels? If so, why the input channel of reference network is three channels without mask channel?

This is really great work, could you please provide the training set?

great work!!! I have a question about the LUTs: "Given one input image, there exist hundreds of LUTs that can be applied to generate its stylized versions and dramatically enrich the training data", is that mean, there are "hundreds of LUTs" templates for you to process the images during the training phase? if yes, where can I get these LUTs? thanks a lot.

Nice work. Would you mind providing training datasets, or sharing the method and details in the collection process?

Looking forward to your reply, thanks.

when run demo.ipynd get output file,I see some images are not normal，like this

I'm not sure what the reason is

i add these code