janusch
janusch
According to @simonbethke there is. I don't know. I simply use colmap :)
This is pretty much against the idea of this project. There won't be an api.
Just because the viewer is written in some web lingo does not make the project attract more contributions. It is called LichtFeld **Studio** because the goal is to become a...
Other idea: https://x.com/hankyang94/status/1922736839410241882 Something like https://poiw.github.io/MODP/index.html or https://repo-sam.inria.fr/nerphys/on-the-fly-nvs/ to create initial poses, applying a round global BA and retraining might be also a quick way to outperform Colmap!
That would require to also port some mono-depth estimation model.
You can do it this way but it is not a proper solution. To implement this, a separate mask folder is required and the masks are applied during training.
https://github.com/MrNeRF/gaussian-splatting-cuda/issues/45 Is prerequisite.
Implement wip: https://github.com/calvin-laurenson/gaussian-splatting-cuda/tree/3dgut
Thx. I promise to review it over the weekend!
@liruilong940607 I think it works quite well as of now but I haven't run the benchmarks yet. What should we do about the file format? In theory, the mlp parameters...