Commit Graph
20 Commits
Author SHA1 Message Date
Abhik Ahuja 11975e9a43 Configurable light direction sampling 2022-11-30 15:14:16 -08:00
Abhik Ahuja b57781ab0c Added dreamfusion to CLI, removed direction embeddings 2022-11-28 14:21:00 -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 a7b2af185c Cleanup configs (#963) 2022-11-15 16:56:56 -08:00
Julian KnodtandMatthew Tancik 9ff167d705 Add temporal distortion and D-NeRF model (#825)
* Add D-NeRF temporal distortion and D-NeRF model

This allows for temporal distortion to be enabled on a variety of models through config
parameters, without an explicit model. Temporal distortion and MIP-NeRF are not currently
jointly runnable but can be changed in the future.

* Simplifying changes

* Simplifying changes

Co-authored-by: Matthew Tancik <tancik@berkeley.edu>
2022-11-15 14:29:29 -08:00
Matthew Tancik ee6b007155 Avoid boolean flag auto conversion (#929) 2022-11-09 20:23:05 -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
Terrance WangandMatthew Tancik 7106f42fcc tensorf (#586)
* tensorf wip

* model changes

* tuning model

* changing imports

* merging

* changing rff encoding to nerfencoding, basis matrix should be properly updated, changed density activation, lr scheduler

* debugging tensorf

* debugging tensorf

* tensorf debugging

* tensorf debugging

* tensorf debugging

* tensorf debug

* debugging tensorf, fixed bug with collider

* tuning model, ray sampling

* adding thresholding for ray computations

* removing unused imports

* making more params configurable in config, adding spherical harmonics

* removing scene contraction, adding arg docstrings

* fixing style

* Fixing bug

Co-authored-by: Matthew Tancik <tancik@berkeley.edu>
2022-11-02 14:32:03 -07:00
Justin Kerr a4755a1349 Sample proposal networks on a schedule to speed training (#770) 2022-10-13 20:14:21 -07:00
Matthew Tancik d24d82d261 Reduce memory usage of nerfacto model (#766)
* Reduce memory usage of nerfacto model

* Tweak config
2022-10-12 18:01:58 -07:00
2150bad0e4 Reducing flicker artifacts for nerfacto method (#750)
* pose opt default nerfacto.

* added SO3 cam pose optimization which is slightly faster (~2.2m/iter) than BARF

* added occupancy grid with placeholder until nerfacc is updated to query density from it

* added occupancy grid sampling as the level 0 in proposal hierarchy, relies on nerfacc 0.1.7 (on pypi)

* merge main

* merge main

* minor changes to sampling

* adjusted sample number

* refactor occ grid stuff, settle on decent params

* nograd proposal sampling

* param tweaks

* varying resolution proposal networks

* params

* merged main

* optimizer update

* remove occ grid

* remove double nerfacto samples

* set single jitter to true

* remove pdb

* more changes

* fix method defaults

* fix pylint

* remove occupancy fn

Co-authored-by: akristoffersen <akristoffersen@berkeley.edu>
Co-authored-by: Justin Kerr <justin.g.kerr@gmail.com>
2022-10-11 14:21:00 -07:00
Ethan Weber 26a41923e5 update camera opt defaults (#732) 2022-10-07 14:59:02 -07:00
Brent Yi 05d264a335 Reduced tyro overhead + helptext formatting (#719)
* `ns-train` helptext formatting for tyro 0.3.23

* Remove unused import
2022-10-06 13:33:42 -07:00
Ethan Weber eaa16e535a max number of iterations logic (#699)
* update max iters

* deal with max iters
2022-10-06 13:07:37 -07:00
Brent Yi edbfccccfe Rename: dcargs -> tyro (#702)
Rename `dcargs` to `tyro`
2022-10-05 22:26:06 -07:00
Brent Yi a8d9c4e0e1 Remove unused rich import (#670) 2022-10-04 21:28:08 -07:00
Brent Yi e9f56e127a Hide fixed args, fix method description formatting (#667) 2022-10-04 21:05:40 -07:00
Matthew Tancik 21f58252c1 remove tensorf model (#660)
* remove tensorf model

* lint
2022-10-04 16:39:47 -07:00
Matthew Tancik 654f070553 Remove mipnerf 360 (#659)
remove mipnerf 360
2022-10-04 16:29:11 -07:00
Ethan Weber 0695879673 [docs] model to method naming (#652)
* renamed model to methods

* name change
2022-10-04 11:53:16 -07:00