Andy

Hi, I also have the same issue, looking forward to your reply!

> I used learning rate 2e-5 for all datasets. I assume that training at 256x256 resolution is unstable with learning rate 1e-4. Thank you, did you use any learning rate...

> I used constant learning rate. I did not try any learning rate scheduler. Thank you very much. BTW, can I ask how many training iterations are needed to produce...

I also have the same problem, I think this part is the most crucial.

> This is a simplified ISP that converts the raw data into sRGB color space. So how does a 14bit data converts to a 16bit data? I noticed that the...

Hello?I have the same question. The raw data mostly is 14-bit(16383). So if you use an factor in[100,300], then the maximum of the simulated low_light raw should be about [55,...

Hi, everybody! I am new in this field, recently I am running a codebase which only gives the dataloader of LLFF. However, I want to try its method for ScanNet,...

> https://trimsh.org/trimesh.visual.html Hi! Could you please explain how do you load a mesh (i.e., an .obj file) into Visuals object using trimesh?

Seems I can manually change .gitignore to avoid it.

@uikino Maybe you could try this repo: https://github.com/CVMI-Lab/UHDM, which can handle 4k color moire images.