Abstract
Recently, Generative Image AI has been applied in product design, but case studies that learn brand identity and apply it to other product groups are still insufficient. Therefore, this study proposes an approach to design new products while maintaining brand identity by using the Generative Image AI model Stable Diffusion. Brand identity was learned from car images of Hyundai Motor and BMW, and then applied to generate designs for scooters, car keys, and robot vacuum cleaners. The results showed that BMW reflected brand identity more effectively than Hyundai, indicating that the effect of image-based learning varies by brand. This study is an attempt to reflect the learning result in the new design based on the image learning ability of the Generative Image AI, and is meaningful in that it presents the possibility as a new design tool to the designer in the future.
| Original language | Korean |
|---|---|
| Pages (from-to) | 121-136 |
| Number of pages | 16 |
| Journal | 디자인융복합연구 |
| Volume | 24 |
| Issue number | 6 |
| DOIs | |
| State | Published - Dec 2025 |
Keywords
- Generative AI
- Brand Identity
- Design Application
- Stable Diffusion
- LoRA
- ControlNet
Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver