Tag
A free interactive book teaching graphics programming with WebGPU in JavaScript, covering from basics to advanced topics like GPU compute and Gaussian splatting.
An immersive 3D tour of Grace Cathedral in San Francisco using Gaussian splatting technique.
This paper proposes PhyMRI-SR, a physics-aware MRI super-resolution method that uses Gaussian splatting and physics-constrained modeling to dynamically adapt resolution-SNR configurations, achieving state-of-the-art performance.
A novel 4D Gaussian splat format called .splat4d with tunable error bounds, offering 16-58x compression over raw splats and native HTTP Range streaming for dynamic scenes, with code and demo available.
CGGS is a text-to-3D framework that improves geometric consistency and quality in ego-centric 3D scene generation using a multi-stage approach with consistency-augmented loss, layout decoration, and geometric refinement via entropy-based depth loss.
This paper introduces CoIn, a novel framework for 3D scene inpainting that bridges 2D diffusion models and 3D Gaussian Splatting via a multi-stage consistency pipeline, enabling both object removal and insertion with flexible masks.
A new approach called Streamable Gaussian Splatting enables real-time streaming of 3D scenes, with a potentially surprising application hinted at with caution.
Lift4D is a test-time optimization framework that reconstructs complete 4D geometry, appearance, and deformation of dynamic objects from a single monocular in-the-wild video, improving over prior methods on challenging sequences with occlusions and non-rigid motion.
FLUX3D introduces a framework for high-fidelity image-to-3D Gaussian Splatting generation by enhancing representation learning and cross-modal alignment with diffusion-aligned structured latents and a sparse-structure-aware diffusion transformer, achieving state-of-the-art results.
SpatialAvatar-0 introduces a multi-stage reconstruction method for high-quality 4D head avatars using a shared FLAME-mesh-bound Gaussian representation, achieving superior performance across benchmarks with reduced iterations.
Apple announced that Gaussian Splatting will be integrated into Apple Maps Flyover this fall, offering photorealistic 3D city views without traditional mesh geometry, derived from their SHARP research model.
At WWDC, Apple showcased impressive Gaussian splatting in Maps, beating Google to the punch with this 3D mapping technology.
UniSHARP extends SHARP for universal monocular view synthesis across diverse camera systems (perspective, fisheye, omnidirectional) by aligning images in an omnidirectional latent space with joint feature and Gaussian space alignment. The method outperforms alternatives on a new benchmark.
Researchers propose Gaussian Point Splatting, a stochastic rendering method using pixel-sized opaque points and 64-bit GPU atomics that scales to hundreds of millions of Gaussians in real time. The method, accepted at SIGGRAPH 2026, employs hierarchical culling and parallel programming primitives to achieve even workload distribution with only minor noise differences compared to original Gaussian splatting.
Tsplat is a Rust-based CLI tool that renders 3D Gaussian Splatting scenes directly in the terminal using only the CPU, even over SSH. It supports Unicode half-blocks and graphics protocols, with controls for navigation and auto-orbit.
4D Gaussian Splatting is a technique that converts flat 2D images into three-dimensional spatial data, enabling reconstruction of different angles from live footage.
The article reports on the application of animated Gaussian splatting technology to create adult content, highlighting a controversial milestone in AI-generated media.
A Gaussian splat of a strawberry, created from 90 perspectives with focus stacking, using slang-splat for training and SuperSplat for viewing.
Real Horizons' Spatial Studio introduces Reframe AI, combining Gaussian splat tours with AI image reimagining to let users pick a 3D view and reimagine it, then optionally bring AI-generated elements back into the spatial experience.
VidSplat is a training-free generative reconstruction framework that uses video diffusion priors to recover complete 3D scenes from sparse inputs by synthesizing novel views.