About your image registration algorithm

[Jian-danai]

I am wondering whether your DRealSR image registration algorithm is the same as the RealSR's?

I mean the iterative registration algorithm from your paper and the supplementary material about your dataset, "To refine the registration results, an image is cropped into patches and an iterative registration algorithm and brightness match are employed". Because I noticed that RealSR also provided an iterative pixel-wise registration algorithm. If not the same, would you mind providing me any more information (such as the mathematical formula) about your registration algorithm?

Thank you so much!!