The progressive growing generative adversarial network is an approach for training a deep convolutional neural network model for generating synthetic images.<\/p>\n
It is an extension of the more traditional GAN architecture that involves incrementally growing the size of the generated image during training, starting with a very small image, such as a 4\u00d74 pixels. This allows the stable training and growth of GAN models capable of generating very large high-quality images, such as images of synthetic celebrity faces with the size of 1024\u00d71024 pixels.<\/p>\n
In this tutorial, you will discover how to develop progressive growing generative adversarial network models from scratch with Keras.<\/p>\n
After completing this tutorial, you will know:<\/p>\n
- \n
- How to develop pre-defined discriminator and generator models at each level of output image growth.<\/li>\n
- How to define composite models for training the generator models via the discriminator models.<\/li>\n
- How to cycle the training of fade-in version and normal versions of models at each level of output image growth.<\/li>\n<\/ul>\n
Discover how to develop DCGANs, conditional GANs, Pix2Pix, CycleGANs, and more with Keras in my new GANs book<\/a>, with 29 step-by-step tutorials and full source code.<\/p>\n
Let\u2019s get started.<\/p>\n
![\"How](\"https:\/\/machinelearningmastery.com\/wp-content\/uploads\/2019\/08\/How-to-Implement-Progressive-Growing-GAN-Models-in-Keras.jpg\")
<\/p>\nHow to Implement Progressive Growing GAN Models in Keras
Photo by Diogo Santos Silva<\/a>, some rights reserved.<\/p>\n<\/div>\nTutorial Overview<\/h2>\n
This tutorial is divided into five parts; they are:<\/p>\n
- \n
- What Is the Progressive Growing GAN Architecture?<\/li>\n
- How to Implement the Progressive Growing GAN Discriminator Model<\/li>\n
- How to Implement the Progressive Growing GAN Generator Model<\/li>\n
- How to Implement Composite Models for Updating the Generator<\/li>\n
- How to Train Discriminator and Generator Models<\/li>\n<\/ol>\n
What Is the Progressive Growing GAN Architecture?<\/h2>\n
GANs are effective at generating crisp synthetic images, although are typically limited in the size of the images that can be generated.<\/p>\n
The Progressive Growing GAN is an extension to the GAN that allows the training of generator models capable of outputting large high-quality images, such as photorealistic faces with the size 1024\u00d71024 pixels. It was described in the 2017 paper by Tero Karras<\/a>, et al. from Nvidia titled \u201cProgressive Growing of GANs for Improved Quality, Stability, and Variation<\/a>.\u201d<\/p>\n
The key innovation of the Progressive Growing GAN is the incremental increase in the size of images output by the generator starting with a 4\u00d74 pixel image and double to 8\u00d78, 16\u00d716, and so on until the desired output resolution.<\/p>\n
\n
Our primary contribution is a training methodology for GANs where we start with low-resolution images, and then progressively increase the resolution by adding layers to the networks.<\/p>\n<\/blockquote>\n
\u2014 Progressive Growing of GANs for Improved Quality, Stability, and Variation<\/a>, 2017.<\/p>\n
This is achieved by a training procedure that involves periods of fine-tuning the model with a given output resolution, and periods of slowly phasing in a new model with a larger resolution.<\/p>\n
\n
When doubling the resolution of the generator (G) and discriminator (D) we fade in the new layers smoothly<\/p>\n<\/blockquote>\n
\u2014 Progressive Growing of GANs for Improved Quality, Stability, and Variation<\/a>, 2017.<\/p>\n
All layers remain trainable during the training process, including existing layers when new layers are added.<\/p>\n
\n
All existing layers in both networks remain trainable throughout the training process.<\/p>\n<\/blockquote>\n
\u2014 Progressive Growing of GANs for Improved Quality, Stability, and Variation<\/a>, 2017.<\/p>\n
Progressive Growing GAN involves using a generator and discriminator model with the same general structure and starting with very small images. During training, new blocks of convolutional layers are systematically added to both the generator model and the discriminator models.<\/p>\n
![\"Example](\"https:\/\/machinelearningmastery.com\/wp-content\/uploads\/2019\/06\/Example-of-Progressively-Adding-Layers-to-Generator-and-Discriminator-Models.-1024x623.png\")
<\/p>\nExample of Progressively Adding Layers to Generator and Discriminator Models.
Taken from: Progressive Growing of GANs for Improved Quality, Stability, and Variation.<\/p>\n<\/div>\nThe incremental addition of the layers allows the models to effectively learn coarse-level detail and later learn ever finer detail, both on the generator and discriminator side.<\/p>\n
\n
This incremental nature allows the training to first discover the large-scale structure of the image distribution and then shift attention to increasingly finer-scale detail, instead of having to learn all scales simultaneously.<\/p>\n<\/blockquote>\n