DCT-Net
DCT-Net copied to clipboard
About the style representation in 3.4.1
Hi, What is the difference between your style representation loss and the surface representation loss of white-box cartoonization?
We followed white-box for style representation loss and no significant difference.
We followed white-box for style representation loss and no significant difference.
As I know, Withe-box Method decomposes images into the surface representation, the structure representation, and the texture representation, which one is the style representation in your paper, or all of them?
all of them are considered.