不确定性驱动的各向异性可见性场三维高斯泼溅主动建图
Uncertainty-driven 3D Gaussian Splatting Active Mapping via Anisotropic Visibility Field
把每个高斯的各向异性可见性场用球谐表示并接入贝叶斯不确定性光栅化器，可在 200 FPS 下为合成视角实时输出不确定性，这一模块可直接复用于任何 3DGS 场景的新视角可靠性评估与主动建图的下一最佳视角选择。
原文依据 we introduce a principled and efficient method for quantifying the visibility field in 3DGS, defined as the anisotropic visibility of each particle with respect to the training views
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文提出高斯泼溅各向异性可见性场（GAVIS），一个用于三维高斯泼溅（3DGS）中不确定性量化与主动建图的新框架。我们的核心洞见是：训练视角未观测到的区域会使 3DGS 产生不可靠的预测。为此，我们引入一种有原则且高效的方法来量化 3DGS 中的可见性场，将其定义为每个粒子相对于训练视角的各向异性可见性，并用球谐函数表示。所得的可见性场被集成到基于贝叶斯网络的不确定性感知 3DGS 光栅化器中，从而能够对合成视角进行实时（200 FPS）的不确定性量化。在此表述之上，我们进一步在最大信息增益框架内执行主动建图。在多种环境中的大量实验表明，GAVIS 在准确性和效率上均持续且显著优于先前方法。此外，除独立使用外，我们的方法还可事后应用于提升现有方法的性能。
英文原文摘要 1079 字符
We present Gaussian Splatting Anisotropic Visibility Field (GAVIS), a novel framework for uncertainty quantification and active mapping in 3DGS. Our key insight is that regions unseen from the training views yield unreliable predictions from the 3DGS. To address this, we introduce a principled and efficient method for quantifying the visibility field in 3DGS, defined as the anisotropic visibility of each particle with respect to the training views, and represented using spherical harmonics. The resulting visibility field is integrated into a Bayesian Network-based uncertainty-aware 3DGS rasterizer, enabling real-time (200 FPS) uncertainty quantification for synthesized views. Active mapping is further performed within a maximum information gain framework building on this formulation. Extensive experiments across diverse environments demonstrate that GAVIS consistently and significantly outperforms prior approaches in both accuracy and efficiency. Moreover, beyond standalone use, our method can be applied post-hoc to improve the performance of existing approaches.