Product Family Formal Design Based on Brand DNA Extraction by AIGC Technology
Contemporary brands increasingly emphasize the unification of design elements to enhance brand image consistency. With the advancement of artificial intelligence, AIGC (Artificial Intelligence Generated Content) technology has been widely applied in the modern design industry. Although existing stud...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11059926/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850086411112284160 |
|---|---|
| author | Yiming Zhang Biyun Wu Jiating Chen Yunkai Chen Mianyu Yu Lufang Zhang Zhichuan Tang |
| author_facet | Yiming Zhang Biyun Wu Jiating Chen Yunkai Chen Mianyu Yu Lufang Zhang Zhichuan Tang |
| author_sort | Yiming Zhang |
| collection | DOAJ |
| description | Contemporary brands increasingly emphasize the unification of design elements to enhance brand image consistency. With the advancement of artificial intelligence, AIGC (Artificial Intelligence Generated Content) technology has been widely applied in the modern design industry. Although existing studies acknowledge the capability of AIGC technology in capturing and preserving brand characteristics, current research has two limitations. Most models rely on machine annotation, and little research addresses how design DNA transfers across different product categories within the same brand using this technology. This study proposes an AIGC-assisted approach for brand product family formal design, aiming to improve the design efficiency and innovation of brand products while maintaining brand consistency across multiple product categories. Through multi-dimensional analysis, design matrix analysis, and fuzzy evaluation methods, this research refines the design DNA of brand product families, providing a theoretical foundation for the subsequent AIGC model training and application. Additionally, based on the Stable Diffusion platform, this study conducts a comparative analysis of LoRA models trained with manual annotations versus CLIP-based machine annotations in inheriting design DNA of brand product families. Research findings suggest that AIGC technology effectively enhances the inheritance of brand attributes across product categories. While both annotation methods demonstrate effectiveness in preserving brand design coherence within ControlNet constraints, the manually labeled LoRA model outperforms its machine-labeled counterpart in representing design elements. This research establishes an innovative approach to brand product design, demonstrating AIGC technology’s practical value in maintaining brand consistency while fostering design innovation. |
| format | Article |
| id | doaj-art-4ee319d98d7245fba06465cf292e6b6c |
| institution | DOAJ |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-4ee319d98d7245fba06465cf292e6b6c2025-08-20T02:43:29ZengIEEEIEEE Access2169-35362025-01-011311355711357210.1109/ACCESS.2025.358465511059926Product Family Formal Design Based on Brand DNA Extraction by AIGC TechnologyYiming Zhang0Biyun Wu1Jiating Chen2Yunkai Chen3https://orcid.org/0009-0005-4786-9867Mianyu Yu4Lufang Zhang5https://orcid.org/0009-0007-3258-3684Zhichuan Tang6https://orcid.org/0000-0002-1730-1120China Tobacco Zhejiang Industrial Company Ltd.,, Hangzhou, ChinaChina Tobacco Zhejiang Industrial Company Ltd.,, Hangzhou, ChinaChina Tobacco Zhejiang Industrial Company Ltd.,, Hangzhou, ChinaSchool of Design and Architecture, Zhejiang University of Technology, Hangzhou, ChinaSchool of Design and Architecture, Zhejiang University of Technology, Hangzhou, ChinaSchool of Design and Architecture, Zhejiang University of Technology, Hangzhou, ChinaSchool of Design and Architecture, Zhejiang University of Technology, Hangzhou, ChinaContemporary brands increasingly emphasize the unification of design elements to enhance brand image consistency. With the advancement of artificial intelligence, AIGC (Artificial Intelligence Generated Content) technology has been widely applied in the modern design industry. Although existing studies acknowledge the capability of AIGC technology in capturing and preserving brand characteristics, current research has two limitations. Most models rely on machine annotation, and little research addresses how design DNA transfers across different product categories within the same brand using this technology. This study proposes an AIGC-assisted approach for brand product family formal design, aiming to improve the design efficiency and innovation of brand products while maintaining brand consistency across multiple product categories. Through multi-dimensional analysis, design matrix analysis, and fuzzy evaluation methods, this research refines the design DNA of brand product families, providing a theoretical foundation for the subsequent AIGC model training and application. Additionally, based on the Stable Diffusion platform, this study conducts a comparative analysis of LoRA models trained with manual annotations versus CLIP-based machine annotations in inheriting design DNA of brand product families. Research findings suggest that AIGC technology effectively enhances the inheritance of brand attributes across product categories. While both annotation methods demonstrate effectiveness in preserving brand design coherence within ControlNet constraints, the manually labeled LoRA model outperforms its machine-labeled counterpart in representing design elements. This research establishes an innovative approach to brand product design, demonstrating AIGC technology’s practical value in maintaining brand consistency while fostering design innovation.https://ieeexplore.ieee.org/document/11059926/Artificial intelligence generated content (AIGC)product family design DNAintelligent designstable diffusionthe LoRA model |
| spellingShingle | Yiming Zhang Biyun Wu Jiating Chen Yunkai Chen Mianyu Yu Lufang Zhang Zhichuan Tang Product Family Formal Design Based on Brand DNA Extraction by AIGC Technology IEEE Access Artificial intelligence generated content (AIGC) product family design DNA intelligent design stable diffusion the LoRA model |
| title | Product Family Formal Design Based on Brand DNA Extraction by AIGC Technology |
| title_full | Product Family Formal Design Based on Brand DNA Extraction by AIGC Technology |
| title_fullStr | Product Family Formal Design Based on Brand DNA Extraction by AIGC Technology |
| title_full_unstemmed | Product Family Formal Design Based on Brand DNA Extraction by AIGC Technology |
| title_short | Product Family Formal Design Based on Brand DNA Extraction by AIGC Technology |
| title_sort | product family formal design based on brand dna extraction by aigc technology |
| topic | Artificial intelligence generated content (AIGC) product family design DNA intelligent design stable diffusion the LoRA model |
| url | https://ieeexplore.ieee.org/document/11059926/ |
| work_keys_str_mv | AT yimingzhang productfamilyformaldesignbasedonbranddnaextractionbyaigctechnology AT biyunwu productfamilyformaldesignbasedonbranddnaextractionbyaigctechnology AT jiatingchen productfamilyformaldesignbasedonbranddnaextractionbyaigctechnology AT yunkaichen productfamilyformaldesignbasedonbranddnaextractionbyaigctechnology AT mianyuyu productfamilyformaldesignbasedonbranddnaextractionbyaigctechnology AT lufangzhang productfamilyformaldesignbasedonbranddnaextractionbyaigctechnology AT zhichuantang productfamilyformaldesignbasedonbranddnaextractionbyaigctechnology |