Style space of latent optimization and mapper

[ssumin6]

Hello
I have read your paper and tested some examples on provided Colab.
It was very interesting, but it left some questions about the style space S and W+.

According to the mentioned paper, it seems that style space S is more disentangled than style space W+.
Is there any reason why you use w+ space instead of S in the latent optimization and mapper method?

[orpatashnik]

Hi @ssumin6 ,
Recently we implemented these two methods in S and found that the results are very similar to those in W+.
I will upload the code this weekend, and also in a newer version of the paper we add these results.

[orpatashnik]

The code is available 😊