A Study of Reinforcement Learning Algorithms for Artistic Creation Guidance in Advertising Design in Virtual Reality Environments

Abstract

Abstract The innovative application of virtual reality technology in advertising design greatly promotes the development of advertising design work in the media field and also provides a new guiding direction for art creation in advertising design. After analyzing the advantages and performance of VR technology applied in advertising design, the article elaborates on the use of art creation rules and related skills in advertising design. To facilitate the guidance of art creation in advertising design within a virtual reality environment, this paper employs chunked feature-matching technology to visually process print advertisement images, thereby providing an interactive design process for advertisements in virtual reality. Then, using a reinforcement learning algorithm as a framework, we combine the DDPG strategy with a neural renderer to construct a virtual reality advertisement design art creation style migration model and conduct design experiments to verify its effectiveness. Finally, we conduct an experience evaluation analysis for the design of virtual reality advertisements. The model in this paper converges after 225 epochs, at which time the loss value is 17.09% lower than that of the CycleGAN model; the mean value of the advertisement image style migration score is 4.08; and the mean value of the comprehensive score of the on-campus test for virtual reality advertisement design is 8.12. Reinforcement learning not only facilitates the art creation guidance of advertisement design in a virtual reality environment but also enhances the user’s understanding of the advertisement design content, thereby laying the foundation for improving the communication effect of virtual advertisements.