A Hierarchical 3D Gaussian Representation for Real-Time Rendering of Very Large Datasets

Abstract

Novel view synthesis has seen major advances in recent years, with 3D Gaussian splatting offering an excellent level of visual quality, fast training and real-time rendering. However, the resources needed for training and rendering inevitably limit the size of the captured scenes that can be represented with good visual quality.

We introduce a hierarchy of 3D Gaussians that preserves visual quality for very large scenes, while offering an efficient LOD solution for efficient rendering of distant content with effective level selection and smooth transitions between levels.

引入了一个3D高斯层次模型，可以为非常大的场景保持视觉质量，同时提供了一个有效的细节层次( LOD )解决方案，通过有效的层次选择和层次之间的平滑过渡，有效地绘制远程内容。

We introduce a divide-and-conquer approach that allows us to train very large scenes in independent chunks. We consolidate the chunks into a hierarchy that can be optimized to further improve visual quality of Gaussians merged into intermediate nodes.

Very large captures typically have sparse coverage of the scene, presenting many challenges to the original 3D Gaussian splatting training method; we adapt and regularize training to account for these issues. We present a complete solution, that enables real-time rendering of very large scenes and can adapt to available resources thanks to our LOD method.

We show results for captured scenes with up to tens of thousands of images with a simple and affordable rig, covering trajectories of up to several kilometers and lasting up to one hour.

Figure

Figure 1

(a) Starting from thousands of calibrated cameras, covering a large area, we subdivide the scene into chunks

(b) We introduce a 3D Gaussian Splatting hierarchy to allow efficient rendering of massive data, that we further optimize to enhance visual quality

(c) We consolidate the hierarchies (d) enabling us to p