Research on Innovative Design of Ink Painting Based on Generative Adversarial Networks

Ink painting has long been regarded as traditional Chinese painting, also known as national painting, and it has long occupied an important position in the history of Chinese painting, but there are many challenges today. This study hopes to explore the innovative design applications of ink painting through GANs techniques. With the continuous advancement of artificial intelligence, GANs have emerged as a technology in visual content creation, demonstrating remarkable capabilities in cross-domain style transfer and innovative aesthetic synthesis across digital art and design fields This thesis analyses the artistic characteristics and cultural connotations of ink painting, and explores its innovative application in the context of the new era by using CycleGAN models. The research hopes to provide a new path for the conversion of ink painting art, and at the same time provides technical support to preserve and inheritance of intangible cultural heritage. Through experimental analyses, the potential and value of style migration based on CycleGAN model in artistic creation is verified.