---
title: "3D Gaussian Splatting"
description: "A technique that reconstructs 3D scenes from 2D images by representing geometry as millions of tiny Gaussian blobs, enabling fast rendering and novel view synthesis."
---

# 3D Gaussian Splatting

**Type:** [What is it?](https://yomu.fyi/visual-explainers/category/what-is-it.md)

![Diagram showing 3D Gaussian Splatting workflow: input images feed into reconstruction, creating a Gaussian point cloud, which renders novel views through differentiable projection.](https://yomufyi.s3.us-east-1.amazonaws.com/visual-explainers/5c74e332-7236-449d-8009-db6c9cee138f/image-72e8d6f7-f3fe-41f8-8719-efc26c8febc6.webp)

> A technique that reconstructs 3D scenes from 2D images by representing geometry as millions of tiny Gaussian blobs, enabling fast rendering and novel view synthesis.

## Explanation

3D Gaussian Splatting is a modern 3D reconstruction method that captures real-world scenes by converting 2D photographs into a dense cloud of Gaussian primitives—small, overlapping spheres with varying opacity and color. Unlike traditional mesh or voxel approaches, Gaussians are lightweight and differentiable, making them trainable end-to-end from image data. The technique excels at rendering novel viewpoints quickly and with high fidelity, which makes it valuable for autonomous vehicle perception, visual effects, robotics, and immersive applications. By preserving the original scene geometry while enabling efficient rendering from new camera angles, Gaussian splatting bridges the gap between photorealism and computational efficiency.

**Tags:** [3D Gaussian Splatting](https://yomu.fyi/topic/3d-gaussian-splatting.md)

## Related reading

### [Scale AV Perception Across Vehicle Platforms with NVIDIA Omniverse NuRec](https://yomu.fyi/post/scale-av-perception-across-vehicle-platforms-with-nvidia-omniverse-nur.md)
- Company: [NVIDIA](https://yomu.fyi/company/nvidia-developer-blog.md)
- Author: Michelle Horton
- Published: Aug 31, 2026
- Type: Tutorial

Autonomous-driving perception changes when the same software moves across carlines because sensor placement, calibration, fields of view, occlusions, geometry, timing, and coverage change. The post presents NVIDIA Omniverse NuRec as a way to adapt a stack before a complete target-carline dataset exists, while retaining real-world data for grounding and validation. NuRec reconstructs recorded drives with 3D Gaussian splatting, renders novel views through a target rig with configurable camera models and calibration, and preserves trajectories, annotations, and map data for label adaptation. The workflow pairs a reconstructed scene with a rig, renders target views, refines frames with NVIDIA Harmonizer, and trains a perception model on the output. It also documents containerized rendering and auxiliary-data generation, including semantic masks, metric depth, and ego-vehicle masks, while noting that views outside observed coverage may be lower quality.