Ten years of generative adversarial nets (GANs): a survey of the state-of-the-art

Abstract

Generative Adversarial Networks (GANs) have rapidly emerged as powerful tools for generating realistic and diverse data across various domains, including computer vision and other applied areas, since their inception in 2014. Consisting of a discriminative network and a generative network engaged in a Minimax game, GANs have revolutionized the field of generative modeling. In February 2018, GAN secured the leading spot on the ``Top Ten Global Breakthrough Technologies List'' issued by the Massachusetts Science and Technology Review. Over the years, numerous advancements have been proposed, leading to a rich array of GAN variants, such as conditional GAN, Wasserstein GAN, CycleGAN, and StyleGAN, among many others. This survey aims to provide a general overview of GANs, summarizing the latent architecture, validation metrics, and application areas of the most widely recognized variants. We also delve into recent theoretical developments, exploring the profound connection between the adversarial principle underlying GAN and Jensen-Shannon divergence while discussing the optimality characteristics of the GAN framework. The efficiency of GAN variants and their model architectures will be evaluated along with training obstacles as well as training solutions. In addition, a detailed discussion will be provided, examining the integration of GANs with newly developed deep learning frameworks such as Transformers, Physics-Informed Neural Networks, Large Language models, and Diffusion models. Finally, we reveal several issues as well as future research outlines in this field.

Review and Experimental Comparison of Generative Adversarial Networks for Synthetic Image Generation

“…After evaluating KID metric (Table 3), it is evident that 1400 epochs are insufficient for both models to generate images of adequate quality, and the improvement in quality stops with continued training. It is hypothesized that such results are achieved due to limited dataset as several studies have found the CGAN and DCGAN models to excel with large-scale datasets such as ImageNet (Chakraborty et al, 2024). In such cases, the application of transfer learning technique, specifically fine-tuning, could be considered.…”