Image de-raining is always an important topic in the computer vision area. In recent years, due to the fast pace development in deep learning, more and more algorithms are developed to perform computer vision tasks. However, few studies are conducted to analyze the current stage of the research. Therefore, in this project, I will briefly review the development of single image de-raining algorithms and give a comprehensive benchmark for three Generative Adversarial Network-based de-raining algorithms.