sketch + style = paints :art: (TOG2018/SIGGRAPH2018ASIA)
Geek Repo:Geek Repo
Github PK Tool:Github PK Tool
Johnkyens opened this issue 5 years ago · comments
在论文中可以看到Unet 有梯度消失的问题,处理简单问题就不会有梯度,论文用了一个组合式的输出图像来优化Loss,那么为什么不用残差网络来解决梯度消失的问题呢😄
It seems that there exists a gradient disappear problem in Unet, why not use a ResNet?