Kwangjin Yoon

Since the paper did not present the parameter settings in detail, I also want to know the secret. In addition to @bsvineethiitg 's five questions, how do you set the...

@bsvineethiitg I implemented the paper unofficially. I saw that noises are very well removed but the sharpness of my results is not comparable to the original results. As you know,...

The author unveiled the training code. Check this out. https://github.com/Tencent/Real-SR After I read the author's implementation, I confirmed that my code was quite close to the author's except for some...

I've re-produced the paper including the traning code. https://github.com/yoon28/unpaired_face_sr

I've re-produced the paper including the training code. https://github.com/yoon28/unpaired_face_sr

what does it mean that "noise could not be eliminated by additive real noise?" Could you explain it in more detail?

No, they don't. Two discriminators don't share the weights.

The dataset, previously, could be downloaded from [here](https://competitions.codalab.org/competitions/22220#participate-get-data), but unfortunately, the download links on the codalab are broken now.

`N_objects = 4` is not a variable, it is a constant. Three objects are moving linearly and one is drawing a circle.

If you want to apply the code to your own video data, I recommend that you should carefully find the parameters for the data, eg, K, Kalman parameters, and so...