2024, Vol. 9, No. 4. - go to content...

Permanent address of this page - https://kostumologiya.ru/en/36tlkl424.html

Метаданные этой статьи так же доступны на русском языке

Full article in PDF format (file size: 1.2 MB)

For citation:

Shakamalova M.D., Ykhin S.S. Development of knitwear patterns using neural networks and modern technologies. Journal of Clothing Science. 2024; 9(4). Available at: https://kostumologiya.ru/PDF/36TLKL424.pdf (in Russian).

Development of knitwear patterns using neural networks and modern technologies

Shakamalova Maria Damirovna
Russian State University named A.N. Kosygin (Technologies. Design. Art), Moscow, Russia
E-mail: Shakamalovamaria@gmail.com

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. The article considers the possibility of using an artificial generative neural network to develop ornaments for light industry products.

In the work, a study was conducted, during which an experiment was performed to generate a series of trial images in the Midjorney neural network based on specified queries, which included generation: associative collages «mood boards», an environment for a podium stage, ornaments for designing knitted fabric. The article demonstrates the results of the generations, which were conducted in English. The source of the creative concept for the design was graphic lines resembling microchips, characteristic of the sci-fi style «cyberpunk».

Based on the images obtained, the author’s collage mood board was compiled using Adobe Photoshop, and a draft design of a jacquard jumper model was developed using Corel Draw.

The work uses the CLO 3D program to perform a virtual fitting of the projected jacquard jumper model. During this process, the model design is developed by applying selected generated ornaments to the details of a digital model scan. Based on the results obtained, it can be concluded that when generating images using prompt text queries using Midjorney’s neural network, combinations of words in English should be used so that the text hints include the object, style guides, input data, and color scheme. Adjustments must be made during the generation to obtain the best results. This ultimately allows for the creation of unique images that have no previously existing analogues.

Keywords: artificial neural networks; artificial intelligence; Midjorney; design engineering; Corel Draw; Adobe Photoshop; CLO 3D; 3D modeling; light industry

Download article in PDF format