Matthew Tancik

Results 260 comments of Matthew Tancik

How to use input equirectangular images for training?

https://user-images.githubusercontent.com/3310961/219978377-f0136119-94b8-4a80-843a-0140df51204e.mp4 There was a bug introduced in nerfstudio 0.1.17 A patch will be coming soon. Alternatively you can use 0.1.16 in the meantime.

How to use input equirectangular images for training?

v0.1.18 is out, use that instead.

Example datasets for depth-nerfacto

Perhaps @mpmisko can share one. He added this functionality.

Export with crop

That should work, can you post the camera path json? Also which version of nerfstudio are you running?

Export with crop

Closing, feel free to reopen if it is still and issue.

OSError: Could not find compatible tinycudann extension for compute capability 61.

Check that your nvidia driver is up to date.

OSError: Could not find compatible tinycudann extension for compute capability 61.

Also try running `pip list | grep torch` The torch version should be `1.12.1+cu113` or `1.12.1+cu116` If the `+cu...` is missing, then you need to reinstall pytorch with cuda support,...

I can't use the "vocab_tree" method to match the images, while it works with the matching way of "sequential"

What error do you see if you run the following in the terminal? ``` colmap vocab_tree_matcher --database_path data\colmap\database.db --SiftMatching.use_gpu 1 --VocabTreeMatching.vocab_tree_path C:\Users\Lemon Zhong\AppData\Local\nerfstudio\nerfstudio\vocab_tree.fbow ```

I can't use the "vocab_tree" method to match the images, while it works with the matching way of "sequential"

Does it work for other data? Can you also try using one of the other matchers, `--matching-method exhaustive`

I can't use the "vocab_tree" method to match the images, while it works with the matching way of "sequential"

Is this also true if you use the data we provide, ie the poster dataset (the data is already processed, but you can run `ns-process-data` on the images folder)? `vocab...