Justin Kerr
Justin Kerr
Closing this since the 1.0 version includes more exact gradients I believe
That would be great, thanks! if there are no gaussians in view it seems reasonable to just return a `torch.full()` with dimensions HxWxC with the background color.
It looks like all of these are warnings, can you show the very end of the message where the error is?
Looks great! thanks for taking this on. Are you planning on adding support for MCMC? it would be nice to remove hyperparameters from SplatfactoConfig and replace them with a sort...
It seems `num_sh_bases` is no longer in the most recent version gsplat, we can fix this import error by manually defining it like this: ``` def num_sh_bases(degree: int) -> int:...
@jb-ye would be good to get some more eyes on this, its very close to ready to merge, we've tested with external methods, train/test/render/export for nerfacto/splatfacto and we're hoping to...
#3569 will fix the failing tests
> Very cool to see this moving along!! Congrats @AntonioMacaronio !! What sort of datasets sizes have you tested so far, like 500 4k images for nerfacto / depth-nerfacto as...
This is great! It would also be good to write documentation or link to installation instructions and the like
Super cool! Would also be great to build out some functionality for previewing the render as it goes and perhaps canceling it. Would be cool to also integrate with viser's...