Artificial Intelligence-Empowered Art Education: A Cycle-Consistency Network-Based Model for Creating the Fusion Works of Tibetan Painting Styles

The integration of Tibetan Thangka and other ethnic painting styles is an important topic of Chinese ethnic art. Its purpose is to explore, supplement, and continue Chinese traditional culture. Restricted by Buddhism and the economy, the traditional Thangka presents the problem of a single style, and drawing a Thangka is time-consuming and labor-intensive. In response to these problems, we propose a Tibetan painting style fusion (TPSF) model based on neural networks that can automatically and quickly integrate the painting styles of the two ethnicities. First, we set up Thangka and Chinese painting datasets as experimental data. Second, we use the training data to train the generator and the discriminator. Then, the TPSF model maps the style of the input image to the target image to fuse the two ethnicities painting styles of Tibetan and Chinese. Finally, to demonstrate the advancement of the proposed method, we add four comparison models to our experiments. At the same time, the Frechet Inception Distance (FID) metric and the questionnaire method were used to evaluate the quality and visual appeal of the generated images, respectively. The experimental results show that the fusion images have excellent quality and great visual appeal.