Authors
Zhengwei Wang, Qi She, Tomas E. Ward
Trinity College Dublin; ByteDance AI Lab
Portals
Summary
In this paper, we review GAN-variants based on performance improvement offered in terms of higher image quality, more diverse images and more stable training. We review the current state of GAN-related research from an architecture and a loss basis.
Abstract
Generative adversarial networks (GANs) have been extensively studied in the past few years. Arguably their most significant impact has been in the area of computer vision where great advances have been made in challenges such as plausible image generation, image-to-image translation, facial attribute manipulation and similar domains. Despite the significant successes achieved to date, applying GANs to real-world problems still poses significant challenges, three of which we focus on here. These are: (1) the generation of high quality images, (2) diversity of image generation, and (3) stable training. Focusing on the degree to which popular GAN technologies have made progress against these challenges, we provide a detailed review of the state of the art in GAN-related research in the published scientific literature. We further structure this review through a convenient taxonomy we have adopted based on variations in GAN architectures and loss functions. While several reviews for GANs have been presented to date, none have considered the status of this field based on their progress towards addressing practical challenges relevant to computer vision. Accordingly, we review and critically discuss the most popular architecture-variant, and loss-variant GANs, for tackling these challenges. Our objective is to provide an overview as well as a critical analysis of the status of GAN research in terms of relevant progress towards important computer vision application requirements. As we do this we also discuss the most compelling applications in computer vision in which GANs have demonstrated considerable success along with some suggestions for future research directions. Code related to GAN-variants studied in this work is summarized on https://github.com/sheqi/GAN_Review.
Contribution
- We focus on GANs by addressing three important problems: (a) High-quality image generation, (b) Diverse image generation, and (c) Stable training
- We propose the novel GAN taxonomy and introduce recent GANs from two perspectives: (a) Architecture of generators and discriminators, e.g., network architecture, latent space, and application driven design, and (b) The objective function for training, e.g., loss design in IPM based and non-IPM based methods, regularization approaches. Compared to other reviews on GANs, this review provides a unique view to different GAN variants
- We also provide the comparison and analysis in terms of pros and cons across GAN-variants presented in this paper