3D Gaussian Splatting
Also known as: 3DGS, Gaussian Splatting
An explicit 3D scene representation, introduced by Kerbl et al. in 2023, in which a scene is modelled as a cloud of anisotropic 3D Gaussians whose positions, shapes, colours, and opacities are optimised to match input views. Unlike NeRFs, which require costly neural-network evaluation per ray, Gaussian splats can be rasterised directly by the GPU, enabling real-time photorealistic rendering on consumer hardware. 3DGS is increasingly used as the underlying representation for AI-assisted scene editing in XR and for accessibility tools that need to query a lightweight 3D model of an environment in real time.
Category: Artificial Intelligence · Machine Learning · Emerging Technology · Immersive Media
Related: Neural Radiance Field · Extended Reality · Virtual reality