Image-Super-Resolution-via-Iterative-Refinement

Image-Super-Resolution-via-Iterative-Refinement copied to clipboard

Why is the loss of Diffusion model calculated between “RANDOM noise” and “model predicted noise”?

4

Thanks for your a lot contribution and hard work. Why is the loss of Diffusion model calculated between “RANDOM noise” and “model predicted noise”? Not between “Actual added noise” and...

egshkim

I'm using your model for academic purposes and I had some issues with your eval.py function until I saw you have a little mistake in line 23. Where you have:...

rmoyav

Poor image quality

9

Thanks for you code, it's a great job. I use python sr.py -p val -c config/sr_sr3_64_512.json and your pretrained models to perform super-resolution. But I get poor results which are...

Zhentao-Liu

Excurl

Train model if my input is 32X32

2

Thanks for the code. I want to know what parameters in the config file must be changed to train a model to upscale a 32X32 image to 256X256.

Shikhar-Soni

SR result on FFHQ is visually much worse than GPEN

1

FYI

anguoyang

Can the model be applied to the improvement of gray image pixels?

1

RyanYip-Kat

Wonderful code! Thanks for your contribution! I wonder if batch size has a large effect on model performance? Or other factors? Looking for your reply!

DongyangHuLi

``` $ conda install -c conda-forge opencv ... The following packages will be UPDATED: pytorch pytorch::pytorch-1.12.1-py3.10_cuda11~ --> pkgs/main::pytorch-1.12.1-cpu_py310h9dbd814_1 ... ```

basab-gupta

How can i transfer a low resolution image to better one? In other words, testing the pretrained model.

2

I read all code and files, I'm not sure whether using infer.py is able to get a ouput from a low resolution image. Like, in srresnet, we can use my...

huchi00057