Content C + Style S = Generated image G
What are Deep ConvNet Learning?
More abstract features in deeper layer.
Content Cost Function
- Say you use hidden layer to compute content cost.
- User pre-trained ConvNet.
- Let and be the activation of layer on the images.
- If and are similar, both images have similar content.
Style Cost Function
- Say you use hidden layer to compute style cost.
- Define style as correlation between activation across different channels.
- Style Matrix :
Let activation at . Let