// HACKER NEWS — CYBERSECURITY
Better Gaussian Splatting in Julia
use WASD + Mouse or tap on the screen to walk around :)
Amulet of St. Michael in Kyiv, Ukraine
(reconstructed with GaussianSplatting.jl, ~6M Gaussians)
GaussianSplatting.jl 2.0
release brings notable quality of life improvements and new capabilities.
Written entirely in Julia it supports following GPU backends:
This is achieved with
KernelAbstractions.jl
that allows writing a single kernel that get's compiled to a specific target.
To avoid freezing the app during heavy work
(such as JIT compilation of GPU kernels, dataset loading, etc.)
the app is now split into two threads:
In this way, the UI always stays responsive and user can interact with it, even if there are long-standing jobs in the background.
It also shows progress bars that something is happening with tips, instead of becoming frozen like before.
UI/UX itself got a big update and is now displays loss plots live during training along with all hyperparameters.
Besides default cloning and splitting densification strategy,
we now support MCMC
(3D Gaussian Splatting as Markov Chain Monte Carlo) densification strategy.
It allows precise control of the number of Gaussians in the scene
and generally relies less on having a good initialization.
Users can select MCMC during dataset loading or in the code with: