改进3D高斯泼溅的稠密化策略，提升重建质量且不增加计算开销。
Improving Densification in 3D Gaussian Splatting for High-Fidelity Rendering
- 基于边缘感知评分选择分裂候选点，提升精度
- 采用长轴分裂减少几何畸变，改善形状还原
- 通过修剪与生长控制缓解过拟合，适合高质量渲染场景
尽管3D高斯泼溅（3DGS）在实时渲染中表现优异，其稠密化策略常导致重建质量不佳。本文从何时分裂、如何分裂及如何缓解过拟合三方面全面优化3DGS的稠密化流程。提出边缘感知评分，有效筛选分裂候选高斯点；引入长轴分裂策略，降低克隆与分裂操作带来的几何畸变；设计恢复感知剪枝、多步更新和生长控制等技术以应对过拟合。所提方法在不增加训练或推理开销的前提下，显著提升渲染保真度，仅用更少高斯点即达到当前最优性能。
原文摘要 · Abstract (English)
Although 3D Gaussian Splatting (3DGS) has achieved impressive performance in real-time rendering, its densification strategy often results in suboptimal reconstruction quality. In this work, we present a comprehensive improvement to the densification pipeline of 3DGS from three perspectives: when to densify, how to densify, and how to mitigate overfitting. Specifically, we propose an Edge-Aware Score to effectively select candidate Gaussians for splitting. We further introduce a Long-Axis Split strategy that reduces geometric distortions introduced by clone and split operations. To address overfitting, we design a set of techniques, including Recovery-Aware Pruning, Multi-step Update, and Growth Control. Our method enhances rendering fidelity without introducing additional training or inference overhead, achieving state-of-the-art performance with fewer Gaussians.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。