[英]Reconstructing a classified image from a simple Convolution Neural Network
I have a CNN trained on a classification task (Network is simple, 2 convolution + pooling layers and 2x fully connected layers). 我有一个接受分类任务训练的CNN(网络很简单,有2个卷积+池化层和2个完全连接的层)。 I would like to use this to reconstruct an image if I input a classification label.
如果输入分类标签,我想用它来重建图像。
Is this possible to achieve? 这有可能实现吗?
Is it possible to share weights between corresponding layers in the 2 networks? 是否可以在两个网络的相应层之间共享权重?
您应该看一下cGAN的实现,以及为什么不从Google获得DeepDream;)答案是yes it is possible
,但并非一帆风顺
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.