Denoising Diffusion Models

Project for Machine Learning for Computer Vision.

In this project we use different PyTorch implementations of diffusion models to perform image denoising. For training we use the Flowers102 dataset.

The concepts behind this project are explained in Ho et al. 2020.

Dependencies

• NumPy
• Matplotlib
• Pandas
• PyTorch

Repository structure

.
├── images    # It contains the images in the notebook     
├── models   # It contains the models    
├── test_pictures   # It contains some test pictures                  
├── utils    # It contains the python files with useful functions (e.g. training utilities)
├── denoising_diffusion_models.ipynb    # Notebook file
├── .gitignore
├── LICENSE
└── README.md

Group members

Name	Surname	Email	Username
Enrico	Pittini	[email protected]	EnricoPittini
Michele	Calvanese	[email protected]	michele98

License

This project is licensed under the MIT License - see the LICENSE file for details.