Image segmentation has become increasingly important in recent years, with applications ranging from autonomous car driving to illness diagnosis. One of the most important tasks in computer vision is image segmentation, which is more difficult than other vision-related tasks since it requires low-level spatial data. In particular, segmentation has greatly benefited from Deep Learning, which has produced a variety of effective models that we use today. Generated Adversarial Networks (GANs), a deep learning technique, have demonstrated impressive results in picture segmentation. The authors of this paper have provided a thorough review analysis of various models Generated Adversarial Networks of and their use in various applications that have been published recently. Three libraries—WoS, PubMed, and Embase (Scopus)—have been taken into consideration for looking up pertinent articles in this field. Out of the 2084 documents found by the search, 52 prospective records were included for final examination after a two-phase screening process. Emerging uses for GANs include the creation of 3D objects, face related detection, image processing, pandemic prevention, texture transfer, and traffic management. Before 2016, there was little study conducted in this area, and after that, applications for it became widely available. The current work also foresees the difficulties with GAN and clears the way for more investigation into this area in the future.