How to train the model with the image2image mode

[dangdinh17]

Thank for your code. I want to train this model with my dataset, and could i use this model for my image, I have a dataset with pairs of original image and noisy image, noisy images have some kind of noises like motion blur, salt and pepper noise and exposure noise for example.

[explainingai-code]

Hi, In this repo, only gaussian noise is added progressively. It seems that based on your requirement, this paper, might be exactly what you are looking for. They have an implementation online - https://github.com/arpitbansal297/Cold-Diffusion-Models , so the deblurring diffusion(with some modifications) should fit your use case.

[dangdinh17]

oh thank you, i will try it