arXiv 2510.085662025-10-09cs.CV

D$^2$GS：深度与密度引导的高斯泼溅用于稳定且准确的稀疏视图重建

D$^2$GS: Depth-and-Density Guided Gaussian Splatting for Stable and Accurate Sparse-View Reconstruction

Insta360 Research（Insta360研究机构）；Tsinghua University（清华大学）；University of California, San Diego（加州大学圣地亚哥分校）；Nanyang Technological University（南洋理工大学）；Wuhan University（武汉大学）

Meixi Song, Xin Lin, Dizhe Zhang, Haodong Li, Xiangtai Li, Bo Du, Lu Qi

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对稀疏视图下3DGS的过拟合与欠拟合问题，提出深度与密度引导的丢弃策略和距离感知保真增强模块，并引入稳定性评估指标，显著提升重建质量与鲁棒性。

摘要：

三维高斯泼溅（3D Gaussian Splatting, 3DGS）的最新进展使得利用显式三维表示实现实时、高保真的新视角合成（Novel View Synthesis, NVS）成为可能。然而，在稀疏视图条件下，性能下降和不稳定性仍然显著。在本工作中，我们识别了稀疏视图条件下的两个关键失败模式：在相机附近高斯密度过大的区域发生过拟合，以及在远处高斯覆盖不足的区域发生欠拟合。为解决这些挑战，我们提出了一个统一框架D$^2$GS，包含两个关键组件：一个深度与密度引导的丢弃（Depth-and-Density Guided Dropout）策略，通过基于密度和深度自适应地掩蔽冗余高斯来抑制过拟合；以及一个距离感知的保真度增强（Distance-Aware Fidelity Enhancement）模块，通过针对性监督提高欠拟合远场区域的重建质量。此外，我们引入了一个新的评估指标来量化所学高斯分布的稳定性，为稀疏视图3DGS的鲁棒性提供见解。在多个数据集上的大量实验表明，我们的方法在稀疏视图条件下显著提高了视觉质量和鲁棒性。项目页面可在以下网址找到：https://insta360-research-team.github.io/DDGS-website/。

英文摘要：

Recent advances in 3D Gaussian Splatting (3DGS) enable real-time, high-fidelity novel view synthesis (NVS) with explicit 3D representations. However, performance degradation and instability remain significant under sparse-view conditions. In this work, we identify two key failure modes under sparse-view conditions: overfitting in regions with excessive Gaussian density near the camera, and underfitting in distant areas with insufficient Gaussian coverage. To address these challenges, we propose a unified framework D$^2$GS, comprising two key components: a Depth-and-Density Guided Dropout strategy that suppresses overfitting by adaptively masking redundant Gaussians based on density and depth, and a Distance-Aware Fidelity Enhancement module that improves reconstruction quality in under-fitted far-field areas through targeted supervision. Moreover, we introduce a new evaluation metric to quantify the stability of learned Gaussian distributions, providing insights into the robustness of the sparse-view 3DGS. Extensive experiments on multiple datasets demonstrate that our method significantly improves both visual quality and robustness under sparse view conditions. The project page can be found at: https://insta360-research-team.github.io/DDGS-website/.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。