NoiseGS: Boosting 3D Gaussian Splatting with Positional Noise for Large-Scale Scene Rendering
Loading...
Date
2025
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association
Abstract
3D Gaussian Splatting (3DGS) efficiently renders 3D spaces by adaptively densifying anisotropic Gaussians from initial points. However, in complex scenes such as city-scale environments, large Gaussians often overlap with high-frequency regions rich in edges and fine details. In these areas, conflicting per-pixel gradient directions cause gradient cancellation, reducing the overall gradient magnitude and potentially causing Gaussians to remain trapped in suboptimal positions even after densification. To address this, we propose NoiseGS, a novel approach that integrates randomized noise injection into 3DGS, guiding suboptimal Gaussians selected for densification toward more optimal positions. In addition, to mitigate the instability caused by oversized Gaussians, we introduce an ℓp-penalization on the scale of Gaussians. Our method integrates seamlessly with existing heuristicbased optimization and demonstrates strong generalization in reconstructing complex scenes such as MatrixCity and Building.
Description
CCS Concepts: Computing methodologies → Point-based models; Computer vision; Machine learning approaches
@inproceedings{10.2312:egs.20251042,
booktitle = {Eurographics 2025 - Short Papers},
editor = {Ceylan, Duygu and Li, Tzu-Mao},
title = {{NoiseGS: Boosting 3D Gaussian Splatting with Positional Noise for Large-Scale Scene Rendering}},
author = {Kweon, Minseong and Cheng, Kai and Chen, Xuejin and Park, Jinsun},
year = {2025},
publisher = {The Eurographics Association},
ISSN = {1017-4656},
ISBN = {978-3-03868-268-4},
DOI = {10.2312/egs.20251042}
}