Re:New Gaussian Splatting viewer that allows code modificati...
by MitPitt on Tuesday July 16, 2024 @11:37AM
(#40977534)
Currently, what's the best way to make gaussian splat scenes from video?
[ Parent
| on HN ]
Re:New Gaussian Splatting viewer that allows code modificati...
by fubei on Tuesday July 16, 2024 @11:45AM
(#40977606)
Thats quite a difficult question, given the large amount of methods (https://github.com/MrNeRF/awesome-3D-gaussian-splatting).
I think "3D Gaussian Splatting as Markov Chain Monte Carlo" looks very promising, as it redesigns the training so that it is no longer dependent on the desification algorithm: https://arxiv.org/abs/2404.09591v1
Re:New Gaussian Splatting viewer that allows code modificati...
by MitPitt on Tuesday July 16, 2024 @11:59AM
(#40977733)
Ah, I'm rather looking for a ready-to-go repo for a layman like me to just have a quick go. The latest I used was nerfstudio a long time ago. Seems to be updated regularly still, I wonder if it supports training splat scenes now. I also wonder if there were any better similar projects since then.
Re:New Gaussian Splatting viewer that allows code modificati...
by was_a_dev on Tuesday July 16, 2024 @12:12PM
(#40977867)
If your video source is of a static scene that the camera is moving through. You could try poly.cam
Upload you video directly to the website for Gaussian splat processing
Re:New Gaussian Splatting viewer that allows code modificati...
by wongarsu on Tuesday July 16, 2024 @02:12PM
(#40978863)
And based on their license acknowledgments poly.cam uses nerfstudio and gsplat, in case anyone else was wondering
Re:New Gaussian Splatting viewer that allows code modificati...
by brizii on Tuesday July 16, 2024 @09:32PM
(#40981861)
Nerfstudio dev here! We have a splatfacto method in nerfstudio that provides a similar interface to our nerf methods.
Re:New Gaussian Splatting viewer that allows code modificati...
by IncreasePosts on Tuesday July 16, 2024 @02:05PM
(#40978798)
That's good - I'm tired of the racial bias that seems to come out of a lot of ML work these days - any algorithm which makes everyone Indian just isn't going to work for most users