1. 09 Feb, 2022 2 commits
  2. 03 Feb, 2022 1 commit
  3. 24 Jan, 2022 4 commits
    • Jeremy Reizenstein's avatar
      use workaround for points_normals · c2862ff4
      Jeremy Reizenstein authored
      Summary:
      Use existing workaround for batched 3x3 symeig because it is faster than torch.symeig.
      
      Added benchmark showing speedup. True = workaround.
      ```
      Benchmark                Avg Time(μs)      Peak Time(μs) Iterations
      --------------------------------------------------------------------------------
      normals_True_3000            16237           17233             31
      normals_True_6000            33028           33391             16
      normals_False_3000        18623069        18623069              1
      normals_False_6000        36535475        36535475              1
      ```
      
      Should help https://github.com/facebookresearch/pytorch3d/issues/988
      
      Reviewed By: nikhilaravi
      
      Differential Revision: D33660585
      
      fbshipit-source-id: d1162b277f5d61ed67e367057a61f25e03888dce
      c2862ff4
    • Jeremy Reizenstein's avatar
      typing for unproject_points · 50531423
      Jeremy Reizenstein authored
      Summary: Fix the base class annotation for unproject_points.
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33281586
      
      fbshipit-source-id: 1c34e8c4b30b359fcb9307507bc778ad3fecf290
      50531423
    • Jeremy Reizenstein's avatar
      avoid deprecated raysamplers · 67778cae
      Jeremy Reizenstein authored
      Summary: Migrate away from NDCGridRaysampler and GridRaysampler to their more flexible replacements.
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33281584
      
      fbshipit-source-id: 65f8702e700a32d38f7cd6bda3924bb1707a0633
      67778cae
    • Jeremy Reizenstein's avatar
      New raysamplers · 3eb42338
      Jeremy Reizenstein authored
      Summary: New MultinomialRaysampler succeeds GridRaysampler bringing masking and subsampling. Correspondingly, NDCMultinomialRaysampler succeeds NDCGridRaysampler.
      
      Reviewed By: nikhilaravi, shapovalov
      
      Differential Revision: D33256897
      
      fbshipit-source-id: cd80ec6f35b110d1d20a75c62f4e889ba8fa5d45
      3eb42338
  4. 21 Jan, 2022 3 commits
    • Jeremy Reizenstein's avatar
      safer pip install in doc · 174738c3
      Jeremy Reizenstein authored
      Summary: Add --no-cache and --no-index to all commands which try to download wheels from S3, to avoid hitting pypi.
      
      Reviewed By: nikhilaravi
      
      Differential Revision: D33507975
      
      fbshipit-source-id: ee796e43cc1864e475cd73c248e9900487012f25
      174738c3
    • Jeremy Reizenstein's avatar
      cameras_from_opencv_projection device #1021 · 45d096e2
      Jeremy Reizenstein authored
      Summary: Fix https://github.com/facebookresearch/pytorch3d/issues/1021 that cameras_from_opencv_projection always creates on CPU.
      
      Reviewed By: nikhilaravi
      
      Differential Revision: D33508211
      
      fbshipit-source-id: fadebd45cacafd633af6a58094cf6f654529992c
      45d096e2
    • Jeremy Reizenstein's avatar
      Join cameras as batch · 39bb2ce0
      Jeremy Reizenstein authored
      Summary:
      Function to join a list of cameras objects into a single batched object.
      
      FB: In the next diff I will remove the `concatenate_cameras` function in implicitron and update the callsites.
      
      Reviewed By: nikhilaravi
      
      Differential Revision: D33198209
      
      fbshipit-source-id: 0c9f5f5df498a0def9dba756c984e6a946618158
      39bb2ce0
  5. 20 Jan, 2022 1 commit
    • Jeremy Reizenstein's avatar
      ambient lights batching #1043 · 9e2bc3a1
      Jeremy Reizenstein authored
      Summary:
      convert_to_tensors_and_broadcast had a special case for a single input, which is not used anywhere except fails to do the right thing if a TensorProperties has only one kwarg. At the moment AmbientLights may be the only way to hit the problem. Fix by removing the special case.
      
      Fixes https://github.com/facebookresearch/pytorch3d/issues/1043
      
      Reviewed By: nikhilaravi
      
      Differential Revision: D33638345
      
      fbshipit-source-id: 7a6695f44242e650504320f73b6da74254d49ac7
      9e2bc3a1
  6. 10 Jan, 2022 2 commits
    • Jeremy Reizenstein's avatar
      drop builds for PyTorch 1.6.0 · fddd6a70
      Jeremy Reizenstein authored
      Summary: PyTorch 1.7.0 was in Oct 2020 and 1.7.1 was in Dec 2020. We shouldn't need older than them, maybe not even 1.7.0.
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33507967
      
      fbshipit-source-id: d3de09c20c44870cbe5522705f2293acc0e62af3
      fddd6a70
    • Jeremy Reizenstein's avatar
      builds for PyTorch 1.10.1 · 85cdcc25
      Jeremy Reizenstein authored
      Summary: Adds 1.10.1 to the nightly builds
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33507966
      
      fbshipit-source-id: af88b155adbc4e3236107f709323bd46a1819013
      85cdcc25
  7. 07 Jan, 2022 1 commit
    • Jeremy Reizenstein's avatar
      initialize pointcloud from list containing Nones · fc4dd802
      Jeremy Reizenstein authored
      Summary:
      The following snippet should work in more cases.
           point_cloud = Pointclouds(
               [pcl.points_packed() for pcl in point_clouds],
               features=[pcl.features_packed() for pcl in point_clouds],
           )
      
      We therefore allow features and normals inputs to be lists which contain some (but not all) Nones.
      
      The initialization of a Pointclouds from empty data is also made a bit better now at working out how many feature channels there are.
      
      Reviewed By: davnov134
      
      Differential Revision: D31795089
      
      fbshipit-source-id: 54bf941ba80672d699ffd5ac28927740e830f8ab
      fc4dd802
  8. 06 Jan, 2022 3 commits
  9. 05 Jan, 2022 1 commit
    • Jeremy Reizenstein's avatar
      remove Python3.6 builds · 49f93b63
      Jeremy Reizenstein authored
      Summary: Python 3.6 was EOL on 2021-12-23.
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33428708
      
      fbshipit-source-id: 37a73898df49a4a49266839278fc8be56597405d
      49f93b63
  10. 04 Jan, 2022 2 commits
    • Jeremy Reizenstein's avatar
      More company name & License · 741777b5
      Jeremy Reizenstein authored
      Summary: Manual adjustments for license changes.
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33405657
      
      fbshipit-source-id: 8a21735726f3aece9f9164da9e3b272b27db8032
      741777b5
    • Jeremy Reizenstein's avatar
      Update license for company name · 9eeb456e
      Jeremy Reizenstein authored
      Summary: Update all FB license strings to the new format.
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33403538
      
      fbshipit-source-id: 97a4596c5c888f3c54f44456dc07e718a387a02c
      9eeb456e
  11. 28 Dec, 2021 1 commit
  12. 21 Dec, 2021 9 commits
    • Nikhila Ravi's avatar
      Update Harmonic embedding in NeRF · 52c71b88
      Nikhila Ravi authored
      Summary: Removed harmonic embedding function from projects/nerf and changed import to be from core pytorch3d.
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33142358
      
      fbshipit-source-id: 3004247d50392dbd04ea72e9cd4bace0dc03606b
      52c71b88
    • Nikhila Ravi's avatar
      Move Harmonic embedding to core pytorch3d · f9a26a22
      Nikhila Ravi authored
      Summary:
      Moved `HarmonicEmbedding` function in core PyTorch3D.
      In the next diff will update the NeRF project.
      
      Reviewed By: bottler
      
      Differential Revision: D32833808
      
      fbshipit-source-id: 0a12ccd1627c0ce024463c796544c91eb8d4d122
      f9a26a22
    • Nikhila Ravi's avatar
      Update use of `select_cameras` · d67662d1
      Nikhila Ravi authored
      Summary: Removed `select_cameras.py` from implicitron and updated all callsites to directly index the cameras.
      
      Reviewed By: bottler
      
      Differential Revision: D33187605
      
      fbshipit-source-id: aaf5b36aef9d72db0c7e89dec519f23646f6aa05
      d67662d1
    • Nikhila Ravi's avatar
      Enable __getitem__ for Cameras to return an instance of Cameras · 28ccdb73
      Nikhila Ravi authored
      Summary:
      Added a custom `__getitem__` method to `CamerasBase` which returns an instance of the appropriate camera instead of the `TensorAccessor` class.
      
      Long term we should deprecate the `TensorAccessor` and the `__getitem__` method on `TensorProperties`
      
      FB: In the next diff I will update the uses of `select_cameras` in implicitron.
      
      Reviewed By: bottler
      
      Differential Revision: D33185885
      
      fbshipit-source-id: c31995d0eb126981e91ba61a6151d5404b263f67
      28ccdb73
    • Jeremy Reizenstein's avatar
      update linux wheel builds · cc3259ba
      Jeremy Reizenstein authored
      Summary:
      * Add PyTorch 1.10 + CUDA 11.1 combination.
      * Change the CUDA 11.3 builds to happen in a separate docker image.
      * Update connection to AWS to use the official `aws` commands instead of the wrapper which is now gone.
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33235489
      
      fbshipit-source-id: 56401f27c002a512ae121b3ec5911d020bfab885
      cc3259ba
    • Jeremy Reizenstein's avatar
      validate sampling_mode · b51be58f
      Jeremy Reizenstein authored
      Summary: New sampling mode option in TexturesUV mush match when collating meshes.
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33235901
      
      fbshipit-source-id: f457473d90bf022e65fe122ef45bf5efad134345
      b51be58f
    • Jeremy Reizenstein's avatar
      force old mistune for website · 74499518
      Jeremy Reizenstein authored
      Summary: When parsing tutorials for building the website, we fix a few libraries at old versions. They need mistune 0.8.4, not the new version 2+, so this gets added to the list of fixed-version libraries.
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33236031
      
      fbshipit-source-id: 2b152b64043edffc59fa909012eab5794c7e8844
      74499518
    • Nikhila Ravi's avatar
      Join points as batch · 262c1bfc
      Nikhila Ravi authored
      Summary: Function to join a list of pointclouds as a batch similar to the corresponding function for Meshes.
      
      Reviewed By: bottler
      
      Differential Revision: D33145906
      
      fbshipit-source-id: 160639ebb5065e4fae1a1aa43117172719f3871b
      262c1bfc
    • Jeremy Reizenstein's avatar
      screen space docstrings fix · eb2bbf84
      Jeremy Reizenstein authored
      Summary: Fix some comments to match the recent change to transform_points_screen.
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33243697
      
      fbshipit-source-id: dc8d182667a9413bca2c2e3657f97b2f7a47c795
      eb2bbf84
  13. 18 Dec, 2021 3 commits
    • Jeremy Reizenstein's avatar
      PyTorch>1.9 version str · 1152a93b
      Jeremy Reizenstein authored
      Summary: Make code for downloading linux wheels robust to double-digit PyTorch minor version.
      
      Reviewed By: nikhilaravi
      
      Differential Revision: D33170562
      
      fbshipit-source-id: 559a97cc98ff8411e235a9f9e29f84b7a400c716
      1152a93b
    • Pyre Bot Jr's avatar
      suppress errors in `vision/fair/pytorch3d` · 315f2487
      Pyre Bot Jr authored
      Differential Revision: D33202801
      
      fbshipit-source-id: d4cb0f4f4a8ad5a6519ce4b8c640e8f96fbeaccb
      315f2487
    • Georgia Gkioxari's avatar
      small fix for iou3d · ccfb72cc
      Georgia Gkioxari authored
      Summary:
      A small numerical fix for IoU for 3D boxes, fixes GH #992
      
      * Adds a check for boxes with zero side areas (invalid boxes)
      * Fixes numerical issue when two boxes have coplanar sides
      
      Reviewed By: nikhilaravi
      
      Differential Revision: D33195691
      
      fbshipit-source-id: 8a34b4d1f1e5ec2edb6d54143930da44bdde0906
      ccfb72cc
  14. 15 Dec, 2021 1 commit
  15. 14 Dec, 2021 2 commits
  16. 13 Dec, 2021 4 commits
    • Jeremy Reizenstein's avatar
      PyTorch 1.10 + CUDA 11.1 builds · d049cd2e
      Jeremy Reizenstein authored
      Summary: Although the PyTorch website, which describes the current version 1.10, suggests CUDA 10.2 and 11.3 are supported, it would appear that we need to include builds for CUDA 11.1 to avoid surprises. This is because these builds are on anaconda, and this combination is used on Google Colab.
      
      Reviewed By: nikhilaravi
      
      Differential Revision: D33063932
      
      fbshipit-source-id: 1b22d1f06e22bd18fb53ceecb58e78ac6a5d1693
      d049cd2e
    • Jeremy Reizenstein's avatar
      version 0.6.1 · 1edc624d
      Jeremy Reizenstein authored
      Summary: Update version number
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33016833
      
      fbshipit-source-id: ee3b0997887ab3bc5779503b13fa2014df41eaed
      1edc624d
    • Jeremy Reizenstein's avatar
      [pytorch3d install.md for 0.6.1 · 6ea63147
      Jeremy Reizenstein authored
      Summary: Update references to dependencies
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33016832
      
      fbshipit-source-id: aa41c7ccc6acd19654303bc18bfd734dc29d88a3
      6ea63147
    • Jeremy Reizenstein's avatar
      update tutorials for release · 093999e7
      Jeremy Reizenstein authored
      Summary: Pre 0.6.1 release, make the tutorials expect wheels with PyTorch 1.10.0
      
      Reviewed By: patricklabatut
      
      Differential Revision: D33016834
      
      fbshipit-source-id: b8c5c1c6158f806c3e55ec668117fa762fa4b75f
      093999e7