arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2409.06407cs.CVcs.LG
收藏

3D场景重建中的不确定性来源

Sources of Uncertainty in 3D Scene Reconstruction

Aalto University(阿尔托大学);University of Oulu(奥卢大学)

Marcus Klasson, Riccardo Mereu, Juho Kannala, Arno Solin

AI总结:

本文梳理NeRF与3D Gaussian Splatting在3D重建中的不确定性来源,并通过学习不确定性输出与集成等方法扩展模型,实证评估其捕捉重建敏感性的能力。

摘要:

3D场景重建过程在真实世界场景中会受到众多不确定性来源的影响。尽管Neural Radiance Fields(NeRF,神经辐射场)和3D Gaussian Splatting(GS,3D高斯泼溅)实现了高保真渲染,但它们缺乏内建机制来直接处理或量化由噪声、遮挡、混杂异常值以及不精确相机位姿输入所引发的不确定性。本文提出一种分类体系,对这些方法中固有的不同不确定性来源进行归类。此外,我们使用不确定性估计技术扩展了基于NeRF和GS的方法,包括学习不确定性输出和集成方法,并开展实证研究,评估它们捕捉重建敏感性的能力。本研究强调,在设计用于具备不确定性感知的3D重建、基于NeRF/GS的方法时,需要处理各种不确定性方面。

英文摘要:

The process of 3D scene reconstruction can be affected by numerous uncertainty sources in real-world scenes. While Neural Radiance Fields (NeRFs) and 3D Gaussian Splatting (GS) achieve high-fidelity rendering, they lack built-in mechanisms to directly address or quantify uncertainties arising from the presence of noise, occlusions, confounding outliers, and imprecise camera pose inputs. In this paper, we introduce a taxonomy that categorizes different sources of uncertainty inherent in these methods. Moreover, we extend NeRF- and GS-based methods with uncertainty estimation techniques, including learning uncertainty outputs and ensembles, and perform an empirical study to assess their ability to capture the sensitivity of the reconstruction. Our study highlights the need for addressing various uncertainty aspects when designing NeRF/GS-based methods for uncertainty-aware 3D reconstruction.

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。

↑