2 ms·

Gaussian splats are like 3 months old. They currently can't be relit, and there's currently no way to turn them into a mesh+texture. If/when either of those pro

Gaussian splats are like 3 months old. They currently can't be relit, and there's currently no way to turn them into a mesh+texture. If/when either of those problems can be solved, I certainly would expect them to be used in the games industry.

magicalhippo 3y ago
> there's currently no way to turn them into a mesh+texture Turning level sets[1] into meshes is old news, like this[2] for example. Given the Gaussians are smooth, I imagine one could readily subtract a fixed threshold for the sum of the Gaussians at any point and get a zero level set that way. Similarly, generating texture coordinates for arbitrary geometry is also well known, and used in texture generation algorithms like this[3]. [1]: https://en.wikipedia.org/wiki/Level-set_method https://en.wikipedia.org/wiki/Level-set_method [2]: https://doc.cgal.org/latest/Surface_mesher/index.html https://doc.cgal.org/latest/Surface_mesher/index.html [3]: https://cvmi-lab.github.io/Point-UV-Diffusion/paper/point_uv_diffusion.pdf https://cvmi-lab.github.io/Point-UV-Diffusion/paper/point_uv...
- blovescoffee 3y ago
  It's theoretically possible to mesh any radiance field. The problem is that the mesh looks like shit.
  - magicalhippo 3y ago
    The gaussian splatting method optimizes the gaussians to lie in the volume of the object, so the approximate the surface well. That's why I figured it might work decently with a level set method. I'm basing this off looking at the examples from the 4D gaussian splatting[1] posted here the other day[2]. [1]: https://guanjunwu.github.io/4dgs/ https://guanjunwu.github.io/4dgs/ [2]: https://news.ycombinator.com/item?id=37905601 https://news.ycombinator.com/item?id=37905601
    - blovescoffee 3y ago
      For certain fine details or certain not-so-fine details like very flat surfaces, gaussian splatting "looks" right but the underlying geometry at the moment is just not correct.

magicalhippo 3y ago

> there's currently no way to turn them into a mesh+texture Turning level sets[1] into meshes is old news, like this[2] for example. Given the Gaussians are smooth, I imagine one could readily subtract a fixed threshold for the sum of the Gaussians at any point and get a zero level set that way. Similarly, generating texture coordinates for arbitrary geometry is also well known, and used in texture generation algorithms like this[3]. [1]: https://en.wikipedia.org/wiki/Level-set_method https://en.wikipedia.org/wiki/Level-set_method [2]: https://doc.cgal.org/latest/Surface_mesher/index.html https://doc.cgal.org/latest/Surface_mesher/index.html [3]: https://cvmi-lab.github.io/Point-UV-Diffusion/paper/point_uv_diffusion.pdf https://cvmi-lab.github.io/Point-UV-Diffusion/paper/point_uv...

blovescoffee 3y ago
It's theoretically possible to mesh any radiance field. The problem is that the mesh looks like shit.
- magicalhippo 3y ago
  The gaussian splatting method optimizes the gaussians to lie in the volume of the object, so the approximate the surface well. That's why I figured it might work decently with a level set method. I'm basing this off looking at the examples from the 4D gaussian splatting[1] posted here the other day[2]. [1]: https://guanjunwu.github.io/4dgs/ https://guanjunwu.github.io/4dgs/ [2]: https://news.ycombinator.com/item?id=37905601 https://news.ycombinator.com/item?id=37905601
  - blovescoffee 3y ago
    For certain fine details or certain not-so-fine details like very flat surfaces, gaussian splatting "looks" right but the underlying geometry at the moment is just not correct.