Cycle-free CycleGAN using Invertible Generator for Unsupervised Low-Dose CT Denoising (IEEE TCI, T. Kwon et al.)
Public dataset that we used was from Low Dose CT Grand Challenge.
You can use train.py for cycle-free CycleGAN.
For train.py, training input & target data, test input & target data folder directory is required.
You can use inference.py for pre-trained cycle-free CycleGAN.
For inference.py, test input & target data folder directory and pre-trained weight file directory is required.