Hey there, could somebody help me? I would like to improve this LoRA: https://civitai.com/models/1828170/foo-fighters-stand-3d-version
Whenever I generate the details look messy and hi res fix can only do so much. Could somebody take a look at it and see what am I doing wrong?
I got the 3D model in Blender but I think that, for some reason, using 3D rendered images in the dataset doesn't work properly...
The training data is attached.