# Software Rasterization of 2 Billion Points in Real Time

arXiv: https://arxiv.org/abs/2204.01287 · HTML: https://mlsanity.com/paper/2204.01287/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Apr 2022.

## Built on by (most cited first, 13 of 13)

- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [A Survey on 3D Gaussian Splatting](https://mlsanity.com/paper/2401.03890/) (2024)
- [RadSplat: Radiance Field-Informed Gaussian Splatting for Robust Real-Time Rendering with 900+ FPS](https://mlsanity.com/paper/2403.13806/) (2024)
- [Gaussian Shell Maps for Efficient 3D Human Generation](https://mlsanity.com/paper/2311.17857/) (2023)
- [TRIPS: Trilinear Point Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2401.06003/) (2024)
- [VR-Splatting: Foveated Radiance Field Rendering via 3D Gaussian Splatting and Neural Points](https://mlsanity.com/paper/2410.17932/) (2024)
- [Faster-GS: Analyzing and Improving Gaussian Splatting Optimization](https://mlsanity.com/paper/2602.09999/) (2026)
- [VET: Visual Error Tomography for Point Cloud Completion and High-Quality Neural Rendering](https://mlsanity.com/paper/2311.04634/) (2023)
- [INPC: Implicit Neural Point Clouds for Radiance Field Rendering](https://mlsanity.com/paper/2403.16862/) (2024)
- [DNF-Avatar: Distilling Neural Fields for Real-time Animatable Avatar Relighting](https://mlsanity.com/paper/2504.10486/) (2025)
- [NARVis: Neural Accelerated Rendering for Real-Time Scientific Point Cloud Visualization](https://mlsanity.com/paper/2407.19097/) (2024)
- [Towards Real-Time Generation of Delay-Compensated Video Feeds for Outdoor Mobile Robot Teleoperation](https://mlsanity.com/paper/2409.09921/) (2024)
- [Rasterized Steered Mixture of Experts for Efficient 2D Image Regression](https://mlsanity.com/paper/2510.05814/) (2025)

## Abstract (arXiv)

We propose a software rasterization pipeline for point clouds that is capable of brute-force rendering up to two billion points in real time (60fps). Improvements over the state of the art are achieved by batching points in a way that a number of batch-level optimizations can be computed before rasterizing the points within the same rendering pass. These optimizations include frustum culling, level-of-detail rendering, and choosing the appropriate coordinate precision for a given batch of points directly within a compute workgroup. Adaptive coordinate precision, in conjunction with visibility buffers, reduces the number of loaded bytes for the majority of points down to 4, thus making our approach several times faster than the bandwidth-limited state of the art. Furthermore, support for LOD rendering makes our software-rasterization approach suitable for rendering arbitrarily large point clouds, and to meet the increased performance demands of virtual reality rendering.