2026, Vol. 11, No. 1. - go to content...

Permanent address of this page - https://kostumologiya.ru/en/20tlkl126.html

Метаданные этой статьи так же доступны на русском языке

Full article in PDF format (file size: 1.1 MB)

For citation:

Demidova M.D., Rykova E.S. Russian generative services as a tool for artistic design of shoe collections. Journal of Clothing Science. 2026; 11(1). Available at: https://kostumologiya.ru/PDF/20TLKL126.pdf (in Russian).

Russian generative services as a tool for artistic design of shoe collections

Demidova Mariia Dmitrievna
Russian State University named A.N. Kosygin (Technologies. Design. Art), Moscow, Russia
E-mail: boorova.maria@ya.ru
ORCID: https://orcid.org/0009-0009-9500-3653
RSCI: https://elibrary.ru/author_profile.asp?id=1239871

Rykova Elena Sergeevna
Russian State University named A.N. Kosygin (Technologies. Design. Art), Moscow, Russia
E-mail: rykova-es@rguk.ru
ORCID: https://orcid.org/0009-0008-3347-0830
RSCI: https://elibrary.ru/author_profile.asp?id=424891

Abstract. This article examines Russian generative services used in artistic shoe design and compares them with international tools. The functionality of Kandinsky, YandexART, and Alice AI Art models is analyzed, including generation parameters, accessibility, text management features, image format support, and the degree of variability of the generated visual material. Using experimental generations based on standardized prompts, the services’ accuracy in interpreting shoe design and technological characteristics, such as model type, material, decorative elements, and proportional parameters, is assessed. It is found that while the models successfully reproduce general stylistic and consumer characteristics, they exhibit instability in the display of design details, half-pair symmetry, hardware elements, and professional terminology, requiring specialist intervention and additional manual adjustments.

Taking into account these identified characteristics, practical scenarios for using generative services during the form search, artistic sketch development, and the generation of variable concepts are considered. Based on empirical data, an algorithm for applying generative models to the design of a single shoe model or collection was developed. This algorithm includes sequential stages of proposal construction, initial generation, image refinement, and design validation. It is demonstrated that such services can serve as an effective auxiliary tool for visual search and expanding the variability of design solutions without replacing the professional competencies of a fashion designer. The practical significance of this study lies in identifying the limitations, potential, and conditions for the application of domestic generative models in the educational and design activities of shoe designers.

Keywords: footwear; art design; sketch design; diffusion models; fashion design; generative artificial intelligence; image generation

Download article in PDF format