423A35C7/pytorch3d
1
0
Fork 0
mirror of https://github.com/facebookresearch/pytorch3d.git synced 2025-12-19 05:40:34 +08:00
Code Issues Packages Projects Releases Wiki Activity

initialize pointcloud from list containing Nones

Browse Source 
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
This commit is contained in:
Jeremy Reizenstein
2022-01-07 05:50:54 -08:00
committed by Facebook GitHub Bot
parent 9640560541
commit fc4dd80208
2 changed files with 105 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

37
tests/test_pointclouds.py
View File

@@ -383,6 +383,43 @@ class TestPointclouds(TestCaseMixin, unittest.TestCase):
self.assertTrue(features_padded[n, p:, :].eq(0).all())
self.assertTrue(points_per_cloud[n] == p)
def test_list_someempty(self):
# We want
# point_cloud = Pointclouds(
# [pcl.points_packed() for pcl in point_clouds],
# features=[pcl.features_packed() for pcl in point_clouds],
# )
# to work if point_clouds is a list of pointclouds with some empty and some not.
points_list = [torch.rand(30, 3), torch.zeros(0, 3)]
features_list = [torch.rand(30, 3), None]
pcls = Pointclouds(points=points_list, features=features_list)
self.assertEqual(len(pcls), 2)
self.assertClose(
pcls.points_padded(),
torch.stack([points_list[0], torch.zeros_like(points_list[0])]),
)
self.assertClose(pcls.points_packed(), points_list[0])
self.assertClose(
pcls.features_padded(),
torch.stack([features_list[0], torch.zeros_like(points_list[0])]),
)
self.assertClose(pcls.features_packed(), features_list[0])
points_list = [torch.zeros(0, 3), torch.rand(30, 3)]
features_list = [None, torch.rand(30, 3)]
pcls = Pointclouds(points=points_list, features=features_list)
self.assertEqual(len(pcls), 2)
self.assertClose(
pcls.points_padded(),
torch.stack([torch.zeros_like(points_list[1]), points_list[1]]),
)
self.assertClose(pcls.points_packed(), points_list[1])
self.assertClose(
pcls.features_padded(),
torch.stack([torch.zeros_like(points_list[1]), features_list[1]]),
)
self.assertClose(pcls.features_packed(), features_list[1])
def test_clone_list(self):
N = 5
clouds = self.init_cloud(N, 100, 5)