Generative adversarial networks in construction applications
Ping Chai, Lei Hou, Guomin Zhang, Quddus Tushar, Yang Zou
Abstract
Generative Adversarial Networks (GANs) have emerged as a powerful tool rapidly advancing the state-of-the-art in numerous domains. This paper conducts a comprehensive review to analyse the applications of GANs in the construction industry over the years, and the review aims to enrich the body of knowledge on this emerging Deep Learning (DL) algorithm in the construction sector. To achieve this, a comprehensive exploration of the variation in GANs is first conducted to establish a general foundation of knowledge. Subsequently, 76 publications from the year 2014 to 2023 are analysed to identify the growth and significance of the current research landscape in the construction field. The results of the study indicate that GANs are predominantly applied in four key construction domains, yet several limitations persist. This study serves as a crucial reference point for researchers, practitioners, and stakeholders seeking to understand and harness the transformative power of GANs in construction.