Learningm

custom_data have so many floaters

1

Hi, How to improve performance when training custom_data? There are so many floaters and blurry part.

Failed to use catacaustics/refnerf related config

2

Hi, I am testing my custom data(with reflective attribute) using llff_360.yaml dataset format, since I use the preprocessing code in LLFF repo to generate poses_bounds.npy file. My scene is object-centric...

Training loss first decrease then increase

4

Hi, thanks for this excellent work. I am trying to finetune vggt on my custom dataset, and i have three questions. 1. I found the training loss first decrease then...