Skip to content

Fine-tuned a Stable Diffusion model to create interior design mockups based on those previously created by the local stakeholder. Completed as part of my ML diploma.

Notifications You must be signed in to change notification settings

KadenShubert/gen-ai-interior-design

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 

Repository files navigation

Stable Diffusion for Interior Design

Problem Statement

An interior design company based in Edmonton aimed to empower clients to visualize their design ideas, regardless of their skill level. Additionally, they aimed to improve efficiency by reducing the time designers spent on creating mockups. Given the creative nature of interior design they explored the use of generative AI. However, it was crucial to ensure that the AI-generated designs could be practically implemented and that the necessary items could be sourced effectively.

Solution

The solution involved fine-tuning a text-to-image Stable Diffusion v1-5 model using DreamBooth and a dataset of previously created mockups labeled with styles and colors for both the room and furnishings. This approach enables both clients and designers to generate new feasible mockups quickly, improving efficiency and client satisfaction.

About

Fine-tuned a Stable Diffusion model to create interior design mockups based on those previously created by the local stakeholder. Completed as part of my ML diploma.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published