the problem of dataset

[algohunt]

the problem of dataset

[AlgoHunt]

Hi, What is your dataset structure?
The default dataset type we use is LLFF. Take flamon_salmon_1 for example the dataset folder should be some thing like this:

flamon_salmon_1
├── 0000    
...
└── 0299 
    ├── pose_bounds.npy  
    └── images
        └── cam[00/01/02/.../20].png

[beibeirory]

Hi, What is your dataset structure? The default dataset type we use is LLFF. Take flamon_salmon_1 for example the dataset folder should be some thing like this:

flamon_salmon_1
├── 0000    
...
└── 0299 
    ├── pose_bounds.npy  
    └── images
        └── cam[00/01/02/.../20].png

Thanks for your reply.
I prepare the dataset folder like "flamon_salmon_1", but the problem exists. I check your code "llff_dataset.py", and find that the dataset folder is like this:
├── pose_bounds.npy

└── images
    └── cam[00/01/02/.../20].png

1、It can not handle different views. Please let me know if there is a problem with my operation.
2、I got another problem.
"Morton code requires a cube grid of power-of-2 size, ignoring...
Segmentation fault (core dumped)"
It looks like there is something wrong with "grid. svox2.SparseGrid()" in "opt.py".

[devinluo27]

Similar problem here @beibeirory ! Just don't know how the authors get through this code and run their experiments.

[Spark001]

Please check the latest commit, I update a script to prepare dataset.