<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>With the increasing demand for personalized fashion, the conventional approach to clothing design struggles to keep up with market expectations. This study explores how artificial intelligence can enhance clothing design, resulting in the creation of a digital customization process that is in step with the evolving trajectory of innovative fashion design. This study integrates the Deeplabv3+ model with a cross-cutting attention mechanism to develop a novel image segmentation network tailored for clothing design, aiming to expand the diversity of design forms. Additionally, the WGAN-GP model is introduced for adaptive optimization of clothing color design, ensuring that the designs align with user preferences. To verify the efficacy of these AI technologies in apparel design, separate simulation verifications for design segmentation and color optimization were conducted. The results show that the Deeplabv3+ network achieved a 6.97% improvement in Mean Intersection over Union (MioU) on the validation dataset, outperforming the OCRNet average by 2.28 percentage points. Moreover, the color optimization with the WGAN-GP model reached a 98.76% color match with the actual garment. Using artificial intelligence technology in apparel design can innovate the design process and provide an adequate technical guarantee to meet the personalized apparel needs of users.</p>
</abstract>

With the increasing demand for personalized fashion, the conventional approach to clothing design struggles to keep up with market expectations. This study explores how artificial intelligence can enhance clothing design, resulting in the creation of a digital customization process that is in step with the evolving trajectory of innovative fashion design. This study integrates the Deeplabv3+ model with a cross-cutting attention mechanism to develop a novel image segmentation network tailored for clothing design, aiming to expand the diversity of design forms. Additionally, the WGAN-GP model is introduced for adaptive optimization of clothing color design, ensuring that the designs align with user preferences. To verify the efficacy of these AI technologies in apparel design, separate simulation verifications for design segmentation and color optimization were conducted. The results show that the Deeplabv3+ network achieved a 6.97% improvement in Mean Intersection over Union (MioU) on the validation dataset, outperforming the OCRNet average by 2.28 percentage points. Moreover, the color optimization with the WGAN-GP model reached a 98.76% color match with the actual garment. Using artificial intelligence technology in apparel design can innovate the design process and provide an adequate technical guarantee to meet the personalized apparel needs of users.

Artificial Intelligence Enabled Apparel Design Research

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

{"article-title":"Artificial Intelligence Enabled Apparel Design Research"}

With the increasing demand for personalized fashion, the conventional approach to clothing design struggles to keep up with market expectations. This study...

Artificial Intelligence Enabled Apparel Design Research

Data publikacji: 22 maj 2024

Zakres stron: -

Otrzymano: 19 lut 2024

Przyjęty: 26 kwi 2024

DOI: https://doi.org/10.2478/amns-2024-1200

Słowa kluczowe
Artificial intelligence, Deeplabv3+ model, Cross-cutting attention, WGAN-GP model, Adaptive optimization, Clothing design

© 2024 Gaolu Huang, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Artificial Intelligence Enabled Apparel Design Research

Data publikacji: 22 maj 2024

Zakres stron: -

Otrzymano: 19 lut 2024

Przyjęty: 26 kwi 2024

DOI: https://doi.org/10.2478/amns-2024-1200

Słowa kluczoweArtificial intelligence, Deeplabv3+ model, Cross-cutting attention, WGAN-GP model, Adaptive optimization, Clothing design

© 2024 Gaolu Huang, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Słowa kluczowe
Artificial intelligence, Deeplabv3+ model, Cross-cutting attention, WGAN-GP model, Adaptive optimization, Clothing design