[论文] Geometry and Gradient-based Partitioning for Panoramic Outdoor Reconst...

论文概要

研究领域: CV 作者: Weijian Chen, Weibo Yao, Yuhang Zhang 发布时间: 2026-07-10 arXiv: 2507.08181

中文摘要

将3D高斯溅射（3DGS）扩展到大型室外场景在数据采集和计算上都代价高昂。采用等距圆柱投影（ERP）的全景图像可通过360°视场减少采集工作量，但由此产生的无处不在的可见性使依赖局部相机视锥的现有分区策略失效，导致块级优化退化为全局训练。因此，我们提出PanoLOG，一种配备几何与梯度分区策略（G²PS）的两阶段由粗到精框架，专为大规模全景3DGS重建设计。在全局粗阶段，PanoLOG利用天球建模和全景单目深度监督获取可靠几何；在精修阶段，G²PS通过视差驱动的不确定性构建自适应包围体，并通过基于梯度的重要性评分分配相机。此外，我们构建了Pano360，首个大型全景室外场景重建基准数据集。大量实验表明，G²PS在保持可扩展的块并行训练的同时，实现了最先进的渲染质量。

原文摘要

Scaling 3D Gaussian Splatting (3DGS) to large outdoor scenes is costly in both data acquisition and computation. Adopting panoramic images with equirectangular projection (ERP) can reduce capture effort via their full 360° field of view, yet the resulting omnipresent visibility invalidates existing partitioning strategies that rely on local camera frustums, causing block-wise optimization to degenerate into global training. Thus, we propose PanoLOG, a two-stage coarse-to-fine framework equipped with a Geometry and Gradient-based Partitioning Strategy tailored for large-scale panoramic 3DGS reconstruction. In the global coarse stage, PanoLOG leverages sky-sphere modeling and panoramic monocular depth supervision for reliable geometry, while in the refinement stage, G²PS builds adaptive boun...

--- *自动采集于 2026-07-11*

#论文 #arXiv #CV #小凯