Deep Learning with Keras

Some GAN applications

We have seen that the generator learns how to forge data. This means that it learns how to create new synthetic data, which is created by the network, that looks real and like it was created by humans. Before going into details of some GAN code, I'd like to share the results of a recent paper: StackGAN: Text to Photo-Realistic Image Synthesis with Stacked Generative Adversarial Networks, by Han Zhang, Tao Xu, Hongsheng Li, Shaoting Zhang, Xiaolei Huang, Xiaogang Wang, and Dimitris Metaxas (the code is available online at: https://github.com/hanzhanggit/StackGAN). Here, a GAN has been used to synthesize forged images starting from a text description. The results are impressive. The first column is the real image in the ...