Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
OpenDAS
nerfacc
Commits
cd72f979
"git@developer.sourcefind.cn:OpenDAS/torch-sparce.git" did not exist on "e8d349ee40d9f86e8c99b5c45ce65989614d4d75"
Commit
cd72f979
authored
Nov 29, 2023
by
Ruilong Li
Browse files
docstring
parent
ad98221d
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
6 additions
and
6 deletions
+6
-6
nerfacc/losses.py
nerfacc/losses.py
+6
-6
No files found.
nerfacc/losses.py
View file @
cd72f979
...
@@ -11,17 +11,17 @@ def distortion(
...
@@ -11,17 +11,17 @@ def distortion(
ray_indices
:
Tensor
,
ray_indices
:
Tensor
,
n_rays
:
int
,
n_rays
:
int
,
)
->
Tensor
:
)
->
Tensor
:
"""Distortion Regularization proposed in Mip-NeRF 360
(on a single GPU)
.
"""Distortion Regularization proposed in Mip-NeRF 360.
Args:
Args:
weights:
[n_samples,] The
weights of the samples.
weights:
The flattened
weights of the samples.
Shape (n_samples,)
t_starts:
[n_samples,]
The start points of the samples.
t_starts: The start points of the samples.
Shape (n_samples,)
t_ends:
[n_samples,]
The end points of the samples.
t_ends: The end points of the samples.
Shape (n_samples,)
ray_indices:
[n_samples,]
The ray indices of the samples.
ray_indices: The ray indices of the samples.
LongTensor with shape (n_samples,)
n_rays: The total number of rays.
n_rays: The total number of rays.
Returns:
Returns:
The per-ray distortion loss with the shape
of [
n_rays, 1
]
.
The per-ray distortion loss with the shape
(
n_rays, 1
)
.
"""
"""
assert
(
assert
(
weights
.
shape
==
t_starts
.
shape
==
t_ends
.
shape
==
ray_indices
.
shape
weights
.
shape
==
t_starts
.
shape
==
t_ends
.
shape
==
ray_indices
.
shape
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment