AI-Powered HTML Automation from Mock-up Images

Abstract

Abstract— To design a website, a developer must first create a webpage, hence a mastery of markup languages is necessary for any engineer. All engineers are expected to design the websites using certain hand-free designs, and then write the code according to the design. This post describes how we combine artificial neural networks, deep learning techniques, and Open CV methodology to create an automated website using these hand-free photos. Our approach also made large datasets available to boost productivity. At the start of a website's design cycle, mock-ups for individual pages are created by hand or with the use of graphic design and specialized mock-up creation technologies. The prototype is subsequently converted into structured HTML or another relevant markup language by software engineers. This method is usually repeated multiple times until the desired template is produced. The objective of this research is to produce code automatically from hand-drawn mockups. Hand-drawn mock-ups are processed using computer vision techniques, and then certain deep learning techniques are employed to develop the proposed system. Our system achieves 73% validation accuracy and 96% method accuracy.