This is a UCU AI project. The project aims to get rid of motion blur on photos. However, ideas that are presented in this work can be applied to many image inpainting tasks. The project presentation lies in deblurGAN_PRO.pdf file.
Team: Nazarii Kuspys, Dmytro Lutchyn
The project supports hydra config. To begin with, one should install requirements.txt. However, you can reinstall pytorch if your working setup differs from (Linux, pip).
To start training the model, you should specify parameters in the config/config.yaml. Also, look through all configurations for the specified parameters (dataset config for example).