This paper aims to compare state-of-the-art change detection techniques like Principal Component Analysis (PCA) with the new advancements inmachine learning techniques like Generative Adversarial Network (GAN). GANis a type of Adversarial Network which uses fake data to discriminate from the real one. This robustness of the GAN makes it ideal for the change detection taskswhich don't have any significant changes. On the other hand, PCA is very effective in finding the significant changes that are profound in the visual analysis of the image. This paper presents both change detection techniques and the other classification methods to present a comparative review of the accuracy and timecomplexities. It was found that the implementation of the GAN-based process ishelpful in specific change detection scenarios with moderate time complexity requirements as compared to PCA

GAN, Change Detection, PCA, comparative metrics