New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Problems about content loss in second pass #25

Open

NikoNairre opened this issue Aug 13, 2024 · 0 comments

NikoNairre commented Aug 13, 2024

The content loss in stage loss is always 0.0000 during training. I found that code in run.py

blend_features_style = vgg(mean_shift(first_pass_img))
content_features = vgg(mean_shift(first_pass_img))
content_loss = content_weight * mse(blend_features_style.relu2_2, content_features.relu2_2)

In this way, these two features are totally identical. What's the correct code to compute content loss for $I_{BR}$ and $I_{B}$?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment