Generative Adversarial Networks: An Overview

TL;DR

This paper provides a comprehensive overview of Generative Adversarial Networks (GANs), explaining their training, applications, and challenges in the signal processing domain.

Contribution

It offers a detailed survey of GAN architectures, training methods, and practical applications, highlighting current challenges and future directions.

Findings

GANs enable learning deep representations without extensive labeled data

They are effective in image synthesis, editing, style transfer, and super-resolution

The paper identifies key challenges in GAN theory and application

Abstract

Generative adversarial networks (GANs) provide a way to learn deep representations without extensively annotated training data. They achieve this through 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 super-resolution and classification. The aim of this review paper 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.