Metal-native differentiable 3D Gaussian Splatting for Apple Silicon: fused forward+backward rasterizer on PyTorch MPS, no CUDA anywhere.
Real-time Gaussian splatting SDKs for Android/Vulkan, iOS/Metal and React Native, sharing a C++17 engine.
3D Gaussian Splatting training on Apple Silicon. Metal kernels that numerically reproduce gsplat — no Python, no CUDA.
Source code for WACV 2026 paper "WarpRF: Multi-View Consistency for Training-Free Uncertainty Quantification and Applications in Radiance Fields"
Video → 3D Gaussian splat. In your browser, or fully local on Apple Silicon. No cloud, no upload.
AURA — Adaptive Unified Radiance Asset: the post-3DGS asset layer. Typed carriers (Gaussian/Beta), calibrated + certified per-carrier confidence exported in KHR_gaussian_splatting & OpenUSD 26.03, unified ray-query, relight preview.
Real-time 3DGS studio for lenticular light-field displays.
RadianceKit — native macOS app for 3D Gaussian Splatting (local, Apple Silicon/Metal). Open docs: export-format reference, capture guide & FAQ.
Capture reality with your phone, explore it as a real-time photorealistic 3D scene — browser-based Gaussian Splatting viewer. WebGL performance engineering, fully client-side.
How precisely can photographs pin down a 3D scene? Camera geometry and minimum Gaussian size in 3D Gaussian Splatting, with reproductions of 3DGS and Mip-Splatting on free GPUs. B.Tech Project, IIT Kharagpur.
Turning a handheld phone video into a 3D Gaussian splat of a real room: three capture runs, the measurements that explain the difference, and the two-line fix that took COLMAP from 2/353 to 351/353 registered images.
React Native component for real-time Gaussian splatting: load an SPZ scene, pick a quality preset, and walk through it from JavaScript.
Native Metal Gaussian splatting SDK for iOS: Swift view, GPU visibility/radix, experimental LOD and hybrid tiles. MIT source and reproducible XCFramework.
DisplayXR demo — real-time 3D Gaussian Splatting viewer for glasses-free 3D displays (OpenXR + Vulkan, Windows & macOS).
Open-vocabulary 3D instance segmentation on Radiant Foam - OpenSplat3D's recipe distilled into Voronoi cells
Drop-in, idempotent speed patches for PGSR surface-reconstruction training — removes per-iteration PCIe transfers, redundant ones-kernel convolutions, and GPU-CPU logging syncs (plus a real EMA-logging bug) with CPU-proven numerical parity: same geometry, lower GPU bill.
Denoising-diffusion RGBD patch prior for regularizing radiance fields — fork of nianticlabs/diffusionerf, used as the diffusion regularizer for few-shot 3DGS.
Few-shot 3D Gaussian Splatting for sparse-view (DTU) reconstruction, using an RGBD patch-diffusion prior (DiffusioNeRF) and PatchmatchNet MVS initialization.
1,421 drone photographs (16.23 GB) of a residential complex in Plymouth, MN, flown as 13 waypoint missions: crosshatched nadir grids + 11 orbits. COLMAP poses for the original photos. CC BY 4.0.
Differentiable Gaussian rasterization with depth & alpha output — fork of graphdeco-inria/diff-gaussian-rasterization, used for few-shot 3DGS.
To associate your repository with the radiance-fields topic, visit your repo's landing page and select "manage topics."