Generate and Segment dental radiography: a comparative study of generative models

This repository aims to implement several generative techniques to both generate and segment dental images. It is organised as follow:

GAN folder

Wasserstein GAN for generation. Modify the parameters in main.py, and run the training using python main.py. You can change the model in model.py, and evaluate the model using eval.py.

baseline folder

diffusion_scratch folder

Finally, split_data.py is a script to split your dataset into train, validation and test sets that you can modify according to your data paths.