Spaces:
Sleeping
Apply for community grant: Academic project (gpu)
I am writing to apply for a GPU grant for our [user-centered residential floor plan prediction and generation] academic project.
Our project is an academic project roughly divided into three parts: residential affordance prediction, floor plan look up and floor plan generation.
The hugging face space serves with the floor plan generation part of the project.
We first created our own floor plan dataset and prompts (Datasets - LuyangZ/FloorAI_18k) based on the open source floor plan dataset (https://doi.org/10.5281/zenodo.7788422).
After that, we trained a ControlNet (Models - LuyangZ/FloorAI), which generates a variety of floor plans by inputting an image of the outline of the apartment as a conditional input and the number of rooms as a prompt.
This tool tries to help architects to quickly generate floor plan references that meet their requirements. However, the limited CPU processing power makes the generation efficiency seriously affected. GPU support is crucial. So we would like to request a GPU grant to provide architects with a more practical and efficient design tool, and to help us maintain and improve the design tool.
Thank you for your time.
