Hi,

Has anyone ever had to noise up the point positions of a .ply sequence and re-export them to be rendered in SIBR?

My workflow has been as follows :

1 - read in a .ply sequence
2 - noise up their point position
3 - write these out as a new .ply sequence using the Gaussian Splat export

As far as I can tell, only the P position gets updated, all other attributes remain the same (f_dc*, f_rest*...).

Is there something I'm missing? Having a really hard time getting this to work.

**Also, I've compared the headers of both files and they seem identical to me.

Any info or guidance would be greatly appreciated!

Thanks!