With the comprehensive advent of the all-media era, the field of digital media art is undergoing unprecedented changes, and the widespread application of computer virtual reality (VR) technology has driven the rapid development of this field. In this context, computer graphics (CG), as one of the core technologies of digital media art design, has become increasingly prominent in enhancing interactive experience. Digital media art design uses images or images as the main medium, and through CG technology, designers can create rich, colorful, and captivating visual works, providing users with immersive interactive experience. This article delves into the practical application of CG in interactive experiences of digital media art and design, and proposes an innovative algorithm that combines the advantages of deep learning (DL) and computer vision (CV) algorithms with CG. Through this algorithm, designers can generate and recognize design elements more intelligently, achieving automation and intelligence in the design process. The experimental results show that the algorithm not only improves design efficiency, but also makes design works more in line with user needs and aesthetic standards.