[andybak]

It's not "rendering" with gaussian splats. It's more "training" (or "fitting"). And not knowing how much the usage vs compute ratio is, I would hesitate to comment.

But knowing a little bit about gaussian splatting, I can't think what manual steps requiring human assistance are even likely to be necessary?

[growthwtf]

Without knowing the specifics of their pipeline I would also hesitate to comment further.

[andybak]

Well we know it's a gaussian splat, we know what the inputs are (RGB and 6dof pose) and we know how gaussian splat training is usually done...

[growthwtf]

Sure, in production envs I have seen humans being used in 3 places: 1. the pose data calibration 2. cleaning up covariances (reducing blobbiness) 3. adding metadata for app usage But, to your point, hard to say which of these or any apply without more info. I would be very very impressed if there were no humans and it's 'just' a training time issue though!