Generating medical images using diffusion

MICCAI 2024 MEC Submission14 Authors

Keywords: diffusion, gerneration

Abstract: In this tutorial, we will walk you through the process of training a diffusion model from scratch using PyTorch and PyTorch Lightning. The model architecture will be based on a UNet-like structure, which is well-suited for capturing complex patterns in images while effectively reversing the diffusion process. We'll also guide you through the setup of a custom image dataset, covering key concepts such as data augmentation, noise schedules, and the overall training loop.

Website: https://colab.research.google.com/drive/1KZSgF1vC0SZygcQOTcFI_vDwYSkTZd8q?usp=sharing

Submission Number: 14