#124 · Primary category: Computer Vision
3dgrut
Ray tracing and hybrid rasterization of Gaussian particles
Project last updated:08/19/26
GitHub Stars
2.4K
Forks
279
Contributors
30
License
Apache-2.0
Why we included this project
Developers and researchers in novel view synthesis will find the reference implementation here for rendering pre-trained 3D Gaussian scenes with a real choice of rendering strategy. The repo pairs two approaches to the same representation. 3DGRT traces rays through volumetric Gaussians, which handles distorted cameras, rolling shutters, and secondary effects like reflections and refraction, but it needs dedicated ray-tracing hardware and runs slower than classic splatting. 3DGUT keeps the same camera and effect support in a rasterization pipeline, so it stays fast on machines without RT cores. A hybrid mode, 3DGRUT, rasterizes primary rays and traces secondary ones. Practical touches are included too: interactive training progress viewers and an exported playground GUI with USD, PLY, and NuRec export. The honest trade-off is the hardware bill, so teams weighing render quality against throughput should plan their GPU budget before adopting the ray-traced path.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
opencv
Open Source Computer Vision Library
RuView
π RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection — all without a single pixel of video.
PaddleOCR
Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.
MinerU
Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
tesseract
Tesseract Open Source OCR Engine (main repository)