Inspiration
We wanted to help universities create custom visuals quickly and easily for events and marketing.
What it does
DiffuseCampus generates customizable, photorealistic images of campus buildings for events and promotions (For now only for UNCC buildings).
How we built it
We fine-tuned the Stable Diffusion XL model with 620 UNCC campus images using DreamBooth.
Challenges we ran into
We had difficulty maintaining architectural accuracy and managing multiple fine-tuned models efficiently.
Accomplishments that we're proud of
We built a platform that produces quality AI images specific to UNCC buildings.
What's next for DiffuseCampus
Next, we’ll add editing features, expand to other schools, and commercialize the platform.
Technical Report
You can read the technical report of our project here -> https://drive.google.com/file/d/1DvLFruVBlkPYoS-2Djl3hm6lT3X4B1fK/view
Built With
- diffusers
- gradio
- huggingface
- python
- stablediffusion
- transformers
Log in or sign up for Devpost to join the conversation.