How can I generate a scene by giving text as input?

tangjiapeng / DiffuScene

[CVPR 2024] DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesis

Other

188 stars 15 forks source link

How can I generate a scene by giving text as input? #24

Open HJJeongInK opened 2 months ago

HJJeongInK commented 2 months ago

I want to create a scene that takes only text as input. However, when I looked at the generate_diffusion.py, there seems to be no codes that takes text as input. There are only a code that saves the text as a txt file. Please let me know. Is it impossible to create a scene by taking text as input?

tangjiapeng commented 2 months ago

The text prompt is generated by the script in the dataset loader. You can change several lines of code to test my pre-trained model on your own text inputs.