zzh-tech/BiT

Questions about the dataset

yzb1997 opened this issue · 3 comments

First of all, congratulations on such a great public work, but I have some questions to ask you:

I would like to use your dataset to train a single image input, single image output deblurring model, I want to know how to find the sharp image corresponding to the blurred image.

Hi, thank you for your interest!
If you want to do one-to-one mapping deblurring problem.
Please use this real-world dataset:
https://github.com/zzh-tech/ESTRNN

Are the two datasets the same?

No, BSD is bigger but can only do one-to-one deblurring.