-
Notifications
You must be signed in to change notification settings - Fork 311
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
add RaDeGS #317
base: main
Are you sure you want to change the base?
add RaDeGS #317
Conversation
Hello, how to test for creating mesh from existing PLY with this PR? |
@@ -400,7 +423,7 @@ def rasterization( | |||
"tile_size": tile_size, | |||
"n_cameras": C, | |||
} | |||
return render_colors, render_alphas, meta | |||
return render_colors, render_alphas, expected_depth, median_depths, expected_normal, meta |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If this is meant to be merged, tests
and the other training scripts in examples
need to be adapted to this API change.
image = imageio.imread(self.parser.image_paths[index])[..., :3] | ||
image = imageio.imread(self.parser.image_paths[index]) | ||
if image.shape[-1]==4: | ||
mask = image[...,-1] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should the conversion to a binary mask be done here? (you're doing data["mask"]/255<0.5
later on)
Any news on this PR? |
mat3x2<T> nJ_inv_T = mat3x2<T>(vv + 1, -uv, // 1st column | ||
-uv, uu + 1, // 2nd column | ||
-u, -v // 3nd column | ||
); | ||
T factor = l / (uu + vv + 1); | ||
vec3<T> Cinv_uvh_n = glm::normalize(Cinv_uvh); | ||
T u_Cinv_u_n_clmap = max(glm::dot(Cinv_uvh_n, uvh), 1E-7); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hello @BaowenZ , I am a little bit lost here and can not align the code with the equation in original paper. Is there any hint?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is related to further derivation of the rendering equation. The plane vector is
Hi, Ruilong is currently writing the code with a similar depth rendering method. And this PR might be helpful as a reference. @ichsan2895 |
hi @BaowenZ can the output of RaDeGS be used in a typical gaussian splat renderer? Or does it require special modifications? |
Appreciate your amazing work! I have a question about the mesh extraction part of the simple_trainer_recon.py.
|
Hi! I added the Rade-GS depth and normal rendering.