2025, Vol. 10, No. 1. - go to content...

Permanent address of this page - https://kostumologiya.ru/en/19tlkl125.html

Метаданные этой статьи так же доступны на русском языке

Full article in PDF format (file size: 786.6 KB)

For citation:

Shakamalova M.D., Ykhin S.S. Designing a digital clothing collection using Neural Networks and 3D-modeling. Journal of Clothing Science. 2025; 10(1). Available at: https://kostumologiya.ru/PDF/19TLKL125.pdf (in Russian).

Designing a digital clothing collection using Neural Networks and 3D-modeling

Shakamalova Maria Damirovna
Russian State University named A.N. Kosygin (Technologies. Design. Art), Moscow, Russia
E-mail: Shakamalovamaria@gmail.com
RSCI: https://elibrary.ru/author_profile.asp?id=1290654

Ykhin Sergey Semennovich
Russian State University named A.N. Kosygin (Technologies. Design. Art), Moscow, Russia
E-mail: sergeyyukhin@yandex.ru
RSCI: https://elibrary.ru/author_profile.asp?id=488023

Abstract. This article explores the application of the generative artificial intelligence model Stable Diffusion in designing patterns for clothing collections, alongside the development of algorithms for prompt engineering to optimize interactions with neural networks. The research aims to analyze innovative approaches to integrating neural networks into fashion design, with a focus on the Stable Diffusion model. The study examines the formulation of structured prompts, which enhance the efficiency and precision of text-to-image generation processes, and demonstrates their role in accelerating design outcomes.

A custom-engineered prompt for text-to-image generation is presented, highlighting its utility in creating fabric prints, textile designs, and decorative elements. A digital clothing collection was developed using the 3D modeling software Style3D, employing an adaptive layout method to integrate AI-generated patterns into garment designs. The practical significance of this work lies in the successful adaptation of AI-generated outputs into a 3D environment, yielding a collection with novel design solutions.

An experiment conducted during the study underscores the effectiveness of combining Stable Diffusion-generated structures with Style3D’s design capabilities, demonstrating their synergy in realizing digital fashion projects. The significance of this research lies in its pioneering integration of cutting-edge technologies with traditional design methodologies, leveraging the visualization of ornamental patterns and clothing concepts to redefine creative workflows.

The findings suggest that neural network-based approaches will increasingly become vital tools for design projects, urging a reevaluation of conventional methodologies in favor of innovative technologies. Such advancements are poised to expand perspectives on the future of fashion design, emphasizing efficiency, creativity, and interdisciplinary collaboration.

Keywords: Stable Diffusion; neural networks; ornament design; 3D modeling; prompt engineering; digital fashion design; generative AI; textile patterns

Download article in PDF format