← All publications
arXiv, 20263DGS

From Splats to Silicon: Rethinking Computational Efficiency of 3DGS

Minnan Pei, Qiwei Dong, Yihan Zhou, Gang Li, Yuchen Zhu, Wenju Zhao, Zhongtian Long, Siting Wang, Peisong Wang and Jian Cheng

Overview

From Splats to Silicon uses GPU profiling to trace when reducing work in 3D Gaussian Splatting actually saves memory traffic or shortens frame time. Rendering new views of a reconstructed scene involves selecting Gaussians, processing their screen-space footprints and moving data between stages. The study compares how those costs change across scenes, viewpoints and implementations. Its companion tool, gsprof, supports investigation of GPU bottlenecks.

三维高斯溅射的计算量减少后,渲染一帧会更快吗?这项研究结合文献和 GPU 性能分析,追踪各阶段的工作量、访存量与耗时。对三维重建后的场景做新视角合成时,同一种优化在不同场景和视角下可能得到不同的帧耗时。论文把高斯筛选、屏幕空间工作量与下游执行联系起来,配套的 gsprof 可用于定位 GPU 瓶颈。

  • 3D Gaussian Splatting
  • Hardware efficiency
  • Profiling

Preprint, 2026. See the linked paper for the full method, experiments, and latest version.

Research by Minnan Pei (裴旻楠) and collaborators · Email