[HN Gopher] Making Gaussian Splats more smaller
       ___________________________________________________________________
        
       Making Gaussian Splats more smaller
        
       Author : ibobev
       Score  : 35 points
       Date   : 2023-09-28 09:10 UTC (13 hours ago)
        
 (HTM) web link (aras-p.info)
 (TXT) w3m dump (aras-p.info)
        
       | samspenc wrote:
       | As someone dabbling in the 3D space as a hobby, Gaussian Splats
       | have been the hot trend these past few weeks and everyone in the
       | 3D space is looking into it for getting realistic photo scans
       | into 3D, as an alternative / supplement to NeRF and
       | photogrammetry.
        
       | imbusy111 wrote:
       | There's probably some locality you can exploit as well. Given the
       | same surface and close proximity, I imagine the coefficients are
       | nearly identical.
       | 
       | Or store only the rotation of certain typical small set of SH
       | coefficients in two values + index. For an ideal, single material
       | sphere, all the coefficients could be lined up exactly, right?
        
         | ReactiveJelly wrote:
         | I think the SH of a splat also depends on geometry around it,
         | for shadows and reflections, so if the sphere is inside a
         | scene, the sphere's SHs may not all be rotations of each other.
         | 
         | Cause I've seen SH used for baked-in shadows.
        
       | Scene_Cast2 wrote:
       | Oh, so that's how Gaussian Splats achieve view-dependence.
       | Strongly reminds me of work on realtime relighting from a decade
       | ago. IIRC some version of Halo used a couple of SH coefficients
       | (either per pixel or per vertex) for some dynamic shadow /
       | lighting thing.
       | 
       | Then, it devolved into figuring out a better set of decomposition
       | functions (SH is Fourier Transform over a spherical coordinate
       | basis). IIRC Haar wavelets worked well too. I'd guess that with
       | modern approaches, a learned function would do even better.
        
       ___________________________________________________________________
       (page generated 2023-09-28 23:00 UTC)