高斯点渲染

高斯点渲染
Gaussian Point Splatting

原始链接: https://momentsingraphics.de/Siggraph2026.html

发表于《ACM图形学汇刊》（SIGGRAPH 2026）的论文《高斯点溅射》（Gaussian Point Splatting）介绍了一种新颖的随机渲染技术，旨在将高斯溅射扩展至包含数亿个元素的大规模场景。作者摒弃了传统的溅射方式，转而从高斯分布中采样像素大小的不透明点，并使用64位原子操作将其渲染到帧缓冲区。该方法利用并行编程原语，将工作负载高效地分配到数百万个GPU线程中。由于线程独立运行，研究人员开发了一种数学形式化方法，以确保不透明度分布的准确性，在保持与原始高斯溅射视觉保真度一致的同时，仅引入了微小的噪声和走样差异。为了进一步提升性能，该方法结合了层级视锥体剔除和遮挡剔除技术，实现了对以往计算成本过高的大规模复杂场景的实时渲染。该技术为新视角合成提供了一种高度可扩展的替代方案，针对现代硬件优化了高密度高斯环境。

Hacker News 关于“高斯点云溅射”（Gaussian Point Splatting）的讨论聚焦于该技术的细微技术差别及潜在的性能瓶颈。评论者分析了该论文的计算需求，指出高斯排序仍然是一个显著的瓶颈，这可能会限制该方法在渲染速度上的提升。关于实现规模，讨论中也有技术澄清：论文中提到的“数百万个线程”指的是 GPU (CUDA) 线程，而非操作系统层面的线程。此外，观察者们将该方法与视觉特效（VFX）行业的现有技术（如点云的随机渐进式绘制）进行了对比。虽然有人认为该方法可能与已有的分块/桶式 GPU 渲染实践重叠，但该成果被 SIGGRAPH 录用，意味着它很可能在领域内做出了新颖且独特的贡献。

Joris Rijsdijk, Christoph Peters, Michael Weinnman, Ricardo Marroquim.
2026–07 in ACM Transactions on Graphics (Proc. SIGGRAPH) 45, 4.
Official version

Abstract

We propose Gaussian point splatting, a stochastic method to render Gaussian splats that scales extremely well to scenes with many Gaussians. Our core idea is to sample pixel-sized, opaque points from the Gaussians and to splat them to a framebuffer using 64-bit atomics. Through parallel programming primitives, we achieve an even distribution of the workload across millions of threads. Since these threads splat points independently, multiple points may splat to the same pixel. That makes it non-trivial to determine how many points should be splatted for a Gaussian or how they should be distributed to achieve the desired opacity. We successfully formalize and solve these problems, thus keeping our renders faithful to the original Gaussian splatting. To further accelerate our method, we employ hierarchical frustum and occlusion culling. Our method renders hundreds of millions of Gaussians in real time. The only differences compared to the original Gaussian splatting are slight noise and differences in aliasing.

Keywords: novel view synthesis, gaussian splatting, large-scale scenes, GPU atomics, point rendering, parallelism

Images

Notes

This work gets presented at SIGGRAPH 2026 on 20th of July. The author's version has been published on 20th of May 2026.