Commit Graph
42 Commits
Author SHA1 Message Date
LSong 9a4f5e8e45 DyCheck dataset parser (#1270)
* DyCheck dataset parser

* fix lint
2023-01-23 13:59:28 -08:00
matsurenandMatthew Tancik 87322113a9 Bug fix for loading cameras with different distortions in nerfstudio_dataparser (#1268)
Co-authored-by: Matthew Tancik <tancik@berkeley.edu>
2023-01-20 14:55:40 -08:00
matsurenandMatthew Tancik 408ab709ec Fix typo (#1231)
* Fix typo COLAMP->COLMAP

* Fix more typo

Co-authored-by: Matthew Tancik <tancik@berkeley.edu>
2023-01-16 12:12:55 -08:00
Michal Pandy 3da877050d Depth supervision for Nerfacto (#1173)
* Implement depth supervision

* Fixes

* Change normalize dtype

* refactor

* Change depth weight

* Small refactor

* Change docstring

* Removu unused import

* Add assumption to depth comment

* Implement URF loss

* Remove default depth loss type param

* URF fixes

* Fix depth loss scale

* Visualize gt depths

* Fixes & documentation

* Documentation and formatting

* Fix formatting problems

* Fix linter problems

* Fix tests

* Fix tests
2023-01-02 10:13:38 +01:00
Matthew Tancik b375727f7e Update docstrings (#1165) 2022-12-23 14:20:48 -07:00
Nikhil Mohan 29df594334 Minimal disk dataparser (#1128)
* wip

* merge upstream

* clkeanup

* cleanup

* cleanup

* cleanup

* more cleanup

* cleanup

* cleanup docs

* lint

* style fix
2022-12-16 20:16:20 +00:00
Matthew Tancik a0a3def162 Fix datamanager dataparser_outputs (#1123)
* Fix datamanager dataparser_outputs

* Cast to float
2022-12-13 13:53:09 -08:00
dozeri83anddror 96f269cc1a instant-ngp-bounded model (#1116)
* inerted a new model configuration instant-ngp-bounded

* added more remarks and small fixes

* went back to old near plane

* added the oprtion to get a file as well as directory in nerfstudio data parser

* pretify prev commit

* review fixes

* more clear explanation in README.md

* reverted readme changes

* fixed merge bug

* review changes. update config.data docs support for json files inputs

Co-authored-by: dror <you@example.com>
2022-12-13 13:05:48 -08:00
Matthew Tancik b41ce8060a Save out dataparser transforms (#1105)
* Save out dataparser transforms

* Make missing folder
2022-12-11 13:02:50 -08:00
ksnzh e605b3344a Fix IndexError if ther is no mask filenames (#1095) 2022-12-05 22:10:55 -08:00
Nicolas ZunhammerandNicolas Zunhammer 6896028fc5 Fix mask filenames misaligned after splitting set into train and eval (#1093)
* Added Dockerfile to build docker image including all dependencies and nerfstudio, added description on how to user docker.

* Move docker description to Installation.md.

* Fix mask_filename indices after splitting train set.

Co-authored-by: Nicolas Zunhammer <n.zunhammer@tum.de>
2022-12-05 17:04:40 -08:00
machenmusik 4c6cd352d8 Make friends_dataparser path separator agnostic (fixes #1066) (#1067)
Make friends_dataparser path separator agnostic
2022-11-30 23:00:25 -08:00
machenmusik 533abb7d2f Fix IndexError runtime error (#1057)
* Fix IndexError runtime error

fixes this error:

  File "C:\code\nerfstudio\nerfstudio\data\dataparsers\nerfstudio_dataparser.py", line 232, in _generate_dataparser_outputs
    distortion_params = torch.stack(distort, dim=0)[idx_tensor]
IndexError: tensors used as indices must be long, byte or bool tensors

* Update nerfstudio_dataparser.py

Per discussion at https://github.com/nerfstudio-project/nerfstudio/pull/1057#issuecomment-1332670064
2022-11-30 15:50:48 -08:00
Matthew Tancik a1f20b3ed8 Add per frame intrinsic support to nerfstudio data (#1049) 2022-11-28 22:13:55 -08:00
Ethan Weber b23d76cc47 Cameras TensorDataclass and Phototourism (#961)
This branch gives us the flexibility to use Cameras as a tensordataclass, giving arbitrary batch dimensions, slicing, etc all for free. It tests this with the phototourism dataset.

Tensordataclass changes:
Originally, the tensordataclass would look at tensors in an object and take everything in front of the last dimension to be the batch dimensions, but in the case of the camera_to_world matrices, everything except for the last TWO are the important dimensions, so we needed to have a way for the tensordataclass to acknowledge these specific fields where there are multiple data dimensions behind the batch dimensions. This is important for things like reshaping where we need to know which dimensions get reshaped and which dimensions get left alone.

Cameras changes:
Now subclass of tensordataclass, which required a change to the tensordataclass itself.
The possibility of extra batch dimensions is the hairy part that requires the majority of these changes. The extra batch dimensions aren't a hassle in every function except in ray generation. The ray generation is where this operates different relative to the old cameras. The reason for this is before, for each ray we wanted to generate, we could specify a single index into our batch of cameras, but if you have multiple camera dimensions, for each ray, you need to now specify one index per camera batch dimension, meaning our camera_indices we specify need to have one extra dimension at the end, a dimension of size Cameras.ndim (number of batch dimensions in the camera object).
I also added some neat logic in generate_rays where if you want your output raybundle to be batched, if you specify batched camera_indices, you will receive the same batch size for the output raybundle, meaning you can easily generate raybundles corresponding to things like image patches and whatnot.

Phototourism changes:
The difficulty here is that images in the phototourism dataset are of different sizes, which is difficult because we can no longer batch the image tensors. We also need to better utilize caching in the CacheDataloader since the vanilladatamanager defaults don't really use it as well as they could, and without properly utilizing it, the rays / sec are maybe 25% of the standard nerfstudio dataset speeds (which are 10% of the size of the phototourism datasets). To do this, we specify a collate function in the existing datamanager configs, and we make a subclass of the vanilladatamanagerconfig named VariableResDataManagerConfig which will have a custom collate function storing the images in a list, and using better defaults for caching in the dataloader, caching ~40 images at a time. The slowest part of using a list is pixel sampling, where we added a custom if case for if our pixel sampler receives a batch of images that are in a list, and the lack of ability to vectorize operations over a list means that caching in the dataloader is even more important.
2022-11-25 22:16:11 -08:00
Matthew Tancik e4b5dea6e4 Fix dataparser filename bug (#1013) 2022-11-23 14:04:07 -08:00
Daniel Crispell 965dcb92cf add support for per-image mask filenames in transforms.json (#957)
* add support for per-image mask filenames in transforms.json

* pass None instead of empty list to DataParserOutputs constructor

* look for downsampled masks in correct folder
2022-11-23 10:18:06 -08:00
Matthew Tancik 532f8067fd Refactor process data (#976) 2022-11-18 09:19:44 -08:00
Matthew Tancik 95a27ecba5 Add support for polycam (#958)
* Add support for polycam

* Modify blur score

* Support processing zip and add undistort fallback

* Created using Colaboratory

* Update docs

* Update demo.ipynb

* Address comments

* fix lint
2022-11-15 17:55:29 -08:00
Zak Murezandzmurez 6ea7dbf934 NuScenes dataparser (#939)
* nuscenes dataparser

* cleanup

* style

* cleanup

* cleanup

* cleanup

Co-authored-by: zmurez <zmurez@eng.ucsd.edu>
2022-11-11 18:40:36 -08:00
Matthew Tancik a13c10ba61 Use tan instead of tanh (#913) 2022-11-06 12:42:30 -08:00
Nikhil Mohan 1a24f3e58c Refactor Data Pipeline (#896)
* add none check

* black formatting

* cleanup

* cleanup

* fix tests

* remove new line

* init

* toy example working

* semantic nerf working

* simplify pixel samplers

* cleanup + fix tests

* comments

* fix tests

* fix tests
2022-11-06 07:40:43 -08:00
Alexander Kristoffersen 0afcd1774b Record3D Dataparser moved to ns-process-data script. (#872)
* init commit.

* added docstring.

* added tyro command.

* bugfix.

* added max dataset size logic.

* string cleanup.

* updated depracation warning.

* typo.
2022-11-03 10:43:52 -07:00
Julian Knodt 35610cb9e2 DNeRF Data Parser (#817)
* Add dnerf dataparser

* Switch time to be stored per camera

I realized it would be better downstream to store times on the ray bundle, and in order to store
them on the ray bundle it's much easier to store them on the camera. Rather than having to
revert this change, I just made it here.
2022-11-02 12:23:13 -07:00
Kevin Chen e8487a0785 Add control for resampling images (#847) 2022-10-31 12:34:13 -07:00
Kevin Chen 359f955f3f Fix skipping non-existent images (#832)
Fix skipping non-existant images
2022-10-26 23:27:08 +03:00
Matthew Tancik 9bed69601f Support python 3.7 (#802)
* Support python 3.7

* Update docs
2022-10-18 16:49:27 -07:00
Matthew Tancik 951d972716 Add more controls for orienting poses (#800) 2022-10-18 13:50:02 -07:00
Ethan WeberandEthan Weber d71c05f256 Google Colab support with viewer (#747)
* allows bridge server to run on a different machine

* added script

* update version

* starting docs

* update docs and scripts

* ngrok support

* use tcp

* moving webrtc outside bridge server. wop

* lazy commit

* lazy commit

* running code. webrtc not working

* lazy commit

* better logic for webrtc connection

* cleanup

* todo comment

* working webrtc

* removed logging

* fixing black

* fix eslint

* fix python version

* fix readme

* misc cleanup

* small docs fix

Co-authored-by: Ethan Weber <ethanweber@Ethans-MacBook-Pro.local>
2022-10-14 15:05:28 -07:00
Alexander Kristoffersen f273d3d9cb Bug fixes on Instant NGP dataparser (#760)
* fix missing attributes for focal length.

* fix scaling.

* dict to tuple

* docstring, error handling.
2022-10-12 10:56:51 -07:00
Ethan Weber 2531242d1d integrating record3d and fix checkpoint delete issue (#722)
* integrating record3d

* fix save checkpoint logic

* added record3d data
2022-10-07 00:34:51 -07:00
Alexander Kristoffersen f885aba702 Record3D Bugfix: Removing Pose OpenGL Conversion (#521)
* added rescale factor.

* merge and remove opengl conversion.

* quick fix.
2022-10-04 19:01:43 -07:00
Matthew Tancik 654f070553 Remove mipnerf 360 (#659)
remove mipnerf 360
2022-10-04 16:29:11 -07:00
Ethan Weber 7ba68dde48 update pipeline docs (#643)
update docs
2022-10-03 23:25:49 -07:00
Ethan Weber 05ba4b6b00 semantic nerfw (#640)
transients not working right now
2022-10-03 20:36:32 -07:00
evonneng 968959bdc9 quickstart updates + (other nits+links added to repo) (#633)
* small changes

* rewrite quickstart + readme

* fixed all licensing headers

* nits
2022-10-03 16:01:16 -07:00
Matt Tancik e1a678e25b Change auto downscale res 2022-10-03 15:38:08 -07:00
Matthew Tancik 6b3746a66a Auto downscale nerfstudio data. (#634) 2022-10-03 14:07:58 -07:00
Ethan Weber 5a76b955cd adding --data alias (#630)
* adding --data alias

* fix black
2022-10-03 12:02:10 -07:00
Matthew Tancik 3efd3c3298 dataset_inputs -> dataparser_outputs (#627) 2022-10-03 09:45:36 -07:00
Matthew Tancik 4110bd7a3e Rename scene bounds (#605) 2022-10-01 20:41:36 -07:00
Ethan Weber e13d401b16 restructure codebase pre release (#602)
* moving a lot

* building
2022-10-01 16:36:25 -07:00