A Review of Screenshot-to-Code Generation Systems

The conversion of screenshots into source code is very important phase in the automation of user interface development. There are artificial-based tools which can understand images and further convert them into executable source code. These system's performance is further improved with the incorporation of generative artificial intelligence. These generative artificial intelligence-based systems achieve high accuracy in converting sketches into source code. Further these systems are helpful for developers, otherwise it was time consuming and costly to convert images into source code and developers often distracted from focusing on logic and functionality. This review paper presents the existing studies which generate source code from sketches using artificial intelligence and machine learning. In this paper, several studies are evaluated and compared based on different parameters. Their strengths and weaknesses are also discussed in this paper. This review paper will help the researchers and authors working in this area to further improve GUI design process and in development of more accurate systems.