Generative Adversarial Networks Projects

VGG loss

The VGG loss is another content loss function, which is applied over generated images and real images. VGG19 is a very popular deep neural network that is mostly used for image classification. VGG19 was introduced by Simonyan and Zisserman in their paper titled Very Deep Convolutional Networks for Large-Scale Image Recognition, which is available at https://arxiv.org/pdf/1409.1556.pdf. The intermediate layers of a pre-trained VGG19 network work as feature extractors and can be used to extract feature maps of the generated images and the real images. The VGG loss is based on these extracted feature maps. It is calculated as the Euclidean distance between the feature maps of the generated image and the real image. The formula for the ...