Generative Adversarial Networks: An Overview

TLDR

Generative adversarial networks learn deep representations without extensive annotated data, enabling applications such as image synthesis, style transfer, superresolution, and classification. This review aims to give signal‑processing researchers an accessible overview of GANs, using familiar analogies and concepts. GANs train two competing networks that generate backpropagation signals, and the review discusses various training and construction methods while highlighting remaining theoretical and practical challenges.

Abstract

Generative adversarial networks (GANs) provide a way to learn deep representations without extensively annotated training data. They achieve this by deriving backpropagation signals through a competitive process involving a pair of networks. The representations that can be learned by GANs may be used in a variety of applications, including image synthesis, semantic image editing, style transfer, image superresolution, and classification. The aim of this review article is to provide an overview of GANs for the signal processing community, drawing on familiar analogies and concepts where possible. In addition to identifying different methods for training and constructing GANs, we also point to remaining challenges in their theory and application.