Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

draft(not working) for mlp color per pixel #123

Draft
wants to merge 2 commits into
base: yp_mlp_color
Choose a base branch
from

Conversation

wanmeihuali
Copy link
Owner

@wanmeihuali wanmeihuali commented Sep 9, 2023

Still not working yet. Please take a look at it first

point_id_list: ti.types.ndarray(ti.i32, ndim=1),
camera_intrinsics: ti.types.ndarray(ti.f32, ndim=2), # (3, 3)
pointcloud: ti.types.ndarray(ti.f32, ndim=2), # (N, 3)
# (N, 56) 56 features (cov_rotation, xxx, r, g, b)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not 56 features after using the MLP. Maybe change to:
#(N, feature_num), feature_num features (cov_rotation, ..., mlp weights and bias)

@@ -534,6 +577,13 @@ def gaussian_point_rasterisation_backward(
):
camera_intrinsics_mat = ti.Matrix(
[[camera_intrinsics[row, col] for col in ti.static(range(3))] for row in ti.static(range(3))])
# K: [fx, 0, cx; 0, fy, cy; 0, 0, 1]
# inv_K: [1/fx, 0, -cx/fx; 0, 1/fy, -cy/fy; 0, 0, 1]
inv_camera_intrinsics_mat = ti.math.mat3([
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

camera_intrinsics_mat and inv_camera_intrinsics_mat seems to be constent vectors. Maybe wrap them in somewhere like CameraInfo to avoid the repeated code and repeated calculation?

[[camera_intrinsics[row, col] for col in ti.static(range(3))] for row in ti.static(range(3))])
# K: [fx, 0, cx; 0, fy, cy; 0, 0, 1]
# inv_K: [1/fx, 0, -cx/fx; 0, 1/fy, -cy/fy; 0, 0, 1]
inv_camera_intrinsics_mat = ti.math.mat3([
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

See comment in the backward func.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Overall lgtm. Some small comments are left.
Looking forward to see the results after updating the place where gaussian_point_rasterisation and gaussian_point_rasterisation_backward are called.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants