Geometry and Gradient-based Partitioning for Panoramic Outdoor Reconstruction
This work addresses the scalability problem of 3D Gaussian Splatting for large outdoor scenes, enabling efficient reconstruction from panoramic images without requiring local camera frustums.
PanoLOG introduces a geometry and gradient-based partitioning strategy (G²PS) for large-scale outdoor 3D reconstruction from panoramic images, achieving state-of-the-art rendering quality with scalable block-parallel training. The method reduces data acquisition cost by using 360° panoramas and avoids degeneration into global training.
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 $360^{\circ}$ 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, G$^2$PS builds adaptive bounding volumes via parallax-driven uncertainty and assigns cameras via gradient-based importance scoring. Furthermore, we construct Pano360, the first benchmark on large-scale panoramic dataset for outdoor scene reconstruction. Extensive experiments demonstrate that G$^2$PS achieves state-of-the-art rendering quality while maintaining scalable, block-parallel training. Our models, training code, and dataset are publicly available.