yeungchenwa / FontDiffuser

[AAAI2024] FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learning
https://yeungchenwa.github.io/fontdiffuser-homepage/
299 stars 26 forks source link

关于StyleEncoder #64

Open inferioronly opened 1 week ago

inferioronly commented 1 week ago

作者您好,感谢贡献出十分出色的工作!关于StyleEncoder的forward中有一点我还不太明白,想请教一下作者,在 for index, blocklist in enumerate(self.blocks): for block in blocklist: h = block(h) if index in self.save_featrues[:-1]: residual_features.append(h) h = self.blocks[-1](h) style_emd = h 这部分代码,这里为什么要让h经过两次self.blocks的last_layer才赋值给style_emd,这样做有什么好处吗? 希望能得到您的回复,不胜感谢!