search

Anttwo / SuGaR

[CVPR 2024] Official PyTorch implementation of SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering
https://anttwo.github.io/sugar/
Other
2.32k stars 175 forks source link

How can I visualize the Gaussians after optimization? #155

Open Kraser42 opened 8 months ago

Kraser42 commented 8 months ago

I truly appreciate your work! Your work is outstanding. I have a question, how can I visualize the Gaussians from a specific perspective after optimization,Just like the images in the leftmost column?

屏幕截图 2024-03-05 195244

KEVIN09876 commented 8 months ago

Same question... Do we have to implement gaussian-shape visualizer ourselves?

Anttwo commented 8 months ago

Hello @Kraser42,

Thank you for your nice words! Please excuse me for my late answer, I've been very busy recently.

To get the leftmost visualization, I did a mesh reconstruction in which I modified the boundig box (see the corresponding arguments for train.py) to reconstruct only a small part of the scene with very high resolution.

I could try to make a small tutorial about it if you're interested!

lxl1115 commented 8 months ago

Hello @Kraser42,

Thank you for your nice words! Please excuse me for my late answer, I've been very busy recently.

To get the leftmost visualization, I did a mesh reconstruction in which I modified the boundig box (see the corresponding arguments for train.py) to reconstruct only a small part of the scene with very high resolution.

I could try to make a small tutorial about it if you're interested!

Could you please make a small tutorial? I am very curious about what the surface of the Gaussians looks like, especially before and after applying SuGaR optimization. It will help me a lot. Thanks so much.

fetian-code commented 7 months ago

Same question. We would really appreciate it if you can make a tutorial about it. Thanks!