Title: Nintendo GAN
Deep Learning techniques have proven to be a very effective tool in the field of computer vision. Most architectures are assigned with the task of classification. As the field of deep learning evolves, these tools have moved beyond classification into applications in image processing. Generative Adversarial Networks (GAN) are a recent development in the effort to increase the data available for training classification networks. These networks are able to produce synthetic data by learning the style and structure of existing data. In this talk we will implement a typical GAN to produce synthetic nintendo screenshots.