* Use AbsGrad to get better results with less gaussians
* some adjustment such that splatfacto-big don't go OOM
* use new gsplat version
---------
Co-authored-by: Jianbo Ye <jianboye@amazon.com>
Co-authored-by: Justin Kerr <justin.g.kerr@gmail.com>
* Update to match viser pr
* Bump viser version
* `done` => `removed` in docs
* Update viewer_elements.py
* Fix
* Point function signature overload
* Update viewer_control.md
* type ignore on callback
---------
Co-authored-by: Brent Yi <yibrenth@gmail.com>
* Support rectangle selection in viewer
* Update doc for click screen_pos
* Raise callback override warning from nerfstudio
* Bump viser version, rect-select only in >0.1.26
* Changed torch version to <=2.2, as <=2.1 will uninstall torch 2.1.2+cu118 and leave users without cuda in torch
* changed <=2.2 to <2.2 to prevent the issue at to prevent the type error referenced here https://github.com/nerfstudio-project/nerfstudio/pull/2857
* Changed torch version to <=2.2, as <=2.1 will uninstall torch 2.1.2+cu118 and leave users without cuda in torch
* Add ability to download EyefulTower dataset
* wip before I copy linning's stuff in
* Generate per-resolution cameras.xml
* Generate transforms.json at download
* Fix a couple of quotes
* Use official EyefulTower splits for train and val
* Disable projectaria-tools on windows
* Fix extra imports
* Add a new nerfacto method tund for EyefulTower
* Split eyefultower download into a separate file
* Add some fisheye support for eyeful data
* Reformatted imports to not be dumb
* Apparently this file was missed when formatting originally
* Added 1k resolution scenes
* revert method_configs.py to original values
* Also add 1k exrs
* Address feedback
* Revert changes to pyproject.toml, to be added in a later PR
* Oops, probably shouldn't have gotten rid of awscli ...
* changed glob variable name
* Update tensor_dataclass.py
---------
Co-authored-by: Ethan Weber <ethanweber@berkeley.edu>
Co-authored-by: Brent Yi <yibrenth@gmail.com>
* undistortion should not occur if pinhole (smoothes the images), and accordingly fix rounding of downscaled image size
* rounding downscaling, more consistent with other resizing methods in public datasets
* reverting to cpu image caching
* re-allow camera unimplemented
* compatibility with mask/depth
* reverted scale factor rounding, corrected mask undistort condition
* moved distortion_params update inside PERSPECTIVE type if-clause
* bumping opencv to 4.8, to deal with undistortion bug with distortion parameters of zero
---------
Co-authored-by: liorf <liorf@amazon.com>
Co-authored-by: alshan <alshan@amazon.com>
Co-authored-by: Justin Kerr <justin.g.kerr@gmail.com>
* Add support for Project Aria datasets.
1) An export script for processing Project Aria datasets
2) Add support for Fisheye624 cameras
3) Add support for center-crop masking and sampling
Example commands for running nerfstudio on Project Aria data:
python scripts/datasets/process_project_aria.py --vrs_file=<path_to_vrs> --mps_data_dir=<path to MPS directory> --output_dir=<nerfstudio_data_dir>
ns-train nerfacto --data <nerfstudio_data_dir> nerfstudio-data --orientation-method none
* Clean up type errors
* Fix formatting
* Fix linter errors, import, and use absolute file paths
* Add aria subcommand to ns-process-data
* Appease ruff
* Add Aria to docs
* Fix docs, add back import try/except
* nit: path.as_posix() -> str(path)
---------
Co-authored-by: Brent Yi <yibrenth@gmail.com>
* Fix large dataset (> 512 images) training in viewer_beta (#2586)
* fix index bug in update_camera_poses of viewer
* Set Ruff line lenth
---------
Co-authored-by: Gina Wu <42229107+ginazhouhuiwu@users.noreply.github.com>
* Add share link support for beta viewer
* More consistent CLI experience
* fix viewer jitter by adding atomic update on camera pose
* change to make_share_url
* fix
* remove incorrect assert
---------
Co-authored-by: Justin Kerr <justin.g.kerr@gmail.com>
