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
2.2k stars 166 forks source link

To pay tribute to SuGaR authors, reproduce SuGaR algorithm in code! #12

Open yuedajiong opened 10 months ago

yuedajiong commented 10 months ago

  1. regularization: guass-points and gauss-point surface-alignment regularization item(s) 'read' the paper line-by-line, char-by-char 'draw' my learning-note as below 'code' the depth-map (the SuGaR author mentioned solution#1, not SuGaR picked solution#2) (speedup^2 paper Eq.9) 'code' the MultivariateNormal sampling (paper Eq.9) (as below) 'code' sampling p 'code' argmin ‘code' all 2023-12-18 In Chinese: 有点担心哪里理解的不对 here: https://github.com/Anttwo/SuGaR/issues/17 DONE

  2. extraction: mesh-extraction by Poisson reconstruction, not Marching-Cubes TODO

  3. joint: joint-optimization both official GS and SuGaR-constraints TODO

clement-chupin commented 10 months ago

Why ? The code is soon available, https://github.com/Anttwo/SuGaR/issues/7#issuecomment-1846470176 You might wait and commit on their code, a bunch of forks from different projects is much harder to merge than commits to the same project 😉

yuedajiong commented 10 months ago

I really need this surface algorithm, and I'm worried in case it's not open source. Another reason is for learning, myself.

yuedajiong commented 10 months ago

progressing ...

image

yuedajiong commented 10 months ago

gauss splat with depth, w/o rgbA (rgb, no A so far, similar with D)

gauss_rasterize.zip

yuedajiong commented 10 months ago

Eq.9 sampling: done!

yuedajiong commented 10 months ago

Eq.3 argmin{}: done!