Research on the Integration Method of Character Costume Culture in Dunhuang Murals Based on Deep Learning Algorithm

Objective. As one of the world’s cultural heritages, Dunhuang murals contain profound cultural, historical, social, and artistic values. In this paper, the evolution and progress of character costumes are analyzed and the development process of Chinese costumes is studied. Methods. This paper compares the characteristics of the costumes in different times; the progress and changes in graphics, styles, and textures; and the different characteristics from the costumes in the modern society. It also compares the coupling degree data of clothing culture in the two periods. Through analysis and research it is found that, compared with traditional algorithms, the integration of clothing culture using deep learning algorithms is more coupling. Through the calculation and analysis of clothing evolution in Dunhuang murals, more design inspiration can be provided for Chinese clothing designers, and more strength and research materials can be contributed to future design and clothing development in combination with ancient and modern clothing culture.