A Review of Generative Adversarial Networks (GANs) and Its Applications in a Wide Variety of Disciplines: From Medical to Remote Sensing

We look into Generative Adversarial Networks (GANs), their prevalent variants, and applications across a number of sectors. GANs combine two neural networks that compete against one another using zero-sum game theory, allowing them to create sharper, more discrete outputs. GANs can be used for image processing, video generation and prediction, and other computer vision applications, as well as science-related tasks such as protein engineering, astronomical data processing, remote sensing image dehazing, and crystal structure synthesis. Other notable fields include finance, marketing, fashion design, sports, and music. This article provides a comprehensive overview of the theory behind GANs, their variants, and metrics used to evaluate them, then presents how GAN and its variants can be applied across twelve domains — from STEM fields such as astronomy and biology, to business fields such as marketing and finance, to arts such as music.

Recommended citation: Dash, A., Ye, J., & Wang, G. (2024). "A Review of Generative Adversarial Networks (GANs) and Its Applications in a Wide Variety of Disciplines: From Medical to Remote Sensing." IEEE Access, 12, 18330–18357.