[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82035-en":3,"doc-seo-82035-105":31,"detail-sidebar-cat-0-en-105":93},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},82035,7971461740909,"Levi","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Geometry and Gradient based Partitioning for Panoramic Outdoor Reconstruction","Scaling 3D Gaussian Splatting (3DGS) to large outdoor environments is limited by expensive data acquisition and heavy computation. Using panoramic images with equirectangular projection (ERP) reduces capture effort thanks to full 360° coverage, but omnipresent visibility breaks frustum-dependent partitioning, turning block-wise optimization into ineffective global training. PanoLOG introduces a two-stage coarse-to-fine framework with a Geometry and Gradient-based Partitioning Strategy (G2PS) plus the Pano360 benchmark to support scalable, high-quality reconstruction.","Geometry and Gradient-based Partitioning for Panoramic Outdoor Reconstruction  \nWeijian Chen 1 ,2 , Weibo Yao 1 ,3 , Yuhang Zhang 1 , Xiaolin Tang 1 , Guo Wang 1 , Weijun Zhang 1 , Xitong Gao4 , Yihao Chen 1 , Hongde Qin5 , Lu Qi 1 ,6 ∗  \n1Insta360 Research, 2 Sun Yat-sen University, 3 South China University of Technology,  \n4 University of Chinese Academy of Sciences, 5 Harbin Engineering University, 6 Wuhan University  \narXiv :2607 .08769v 1 [ cs .CV] 9 Jul 2026  \nFigure 1: Overview of PanoLOG, which enables scalable large-scale outdoor reconstruction by introducing G2PS, a Geometry and Gradient-based Partitioning Strategy for 360◦ panoramic scenes.  \nAbstract  \nScaling 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 (G2PS) tailored for large-scale panoramic 3DGS reconstruction. In the global coarse stage,  \n∗Corresponding author: Lu Qi  \nPreprint.  \nPanoLOG leverages sky-sphere modeling and panoramic monocular depth supervision for reliable geometry, while in the refinement stage, G2PS builds adaptive bounding volumes via parallax-driven uncertainty and assigns cameras via gradientbased importance scoring. Furthermore, we construct Pano360, the first benchmark on large-scale panoramic dataset for outdoor scene reconstruction. Extensive experiments demonstrate that G2PS achieves state-of-the-art rendering quality while maintaining scalable, block-parallel training. Our models, training code, and dataset are publicly available [https://insta360-research-team.github.io/GGPS](https://insta360-research-team.github.io/GGPS)Website/ .  \n1 Introduction  \nNovel View Synthesis [Chung et al., 2025, Wu et al., 2025, 2024] has emerged as a cornerstone of computer vision, underpinning applications from autonomous driving simulation to virtual reality. Recently, 3D Gaussian Splatting (3DGS) has become the mainstream paradigm for photorealistic reconstruction, leveraging anisotropic explicit Gaussian primitives and an efficient differentiable rasterization pipeline to achieve unprecedented rendering speed.  \nDespite its success, scaling 3DGS to expansive outdoor environments remains challenging due to the prohibitive costs of data acquisition and computational overhead. Conventional pipelines typically rely on capturing a massive volume of narrow field-of-view (FoV) images, subsequently adopting a divide-and-conquer strategy to partition the scene. However, such workflows are not only labor-intensive but also prone to structural inconsistencies across stitching boundaries, limiting their efficiency for large-scale deployment.  \nInspired by omnidirectional vision, adopting a panoramic ERP representation [Lin et al., 2025, Ge et al., 2026, Feng et al., 2026] offers a compelling alternative by capturing a full 360◦ field of view in a single shot, thereby streamlining data collection. Yet, this shift moves the complexity from acquisition to spatial partitioning. Traditional partitioning [Kerbl et al., 2024, Liu et al., 2024a, Lin et al., 2024] relies on the local visibility of pinhole cameras, which is at odds with the 360◦ nature of panoramic vision. Without frustum constraints, the omnipresent visibility in panoramic scenes strips these strategies of their discriminative power, causing intended block-wise partitioning to degenerate back into inefficient global optimization. Thus, one question raised: could we have a specific partition strategy that can adapt for panoramic images?  \nTo address this issue, we propose G2PS, a two-stage coarse-to-fine tr","cbCaid4GMOZRfiYr","https://ap.wps.com/l/cbCaid4GMOZRfiYr","pdf",16647449,6,1,16,"English","en",105,"# Introduction\n## Motivation and Challenges\n## Proposed Approach (G2PS and PanoLOG)\n## Benchmark and Contributions","[{\"question\":\"Why do existing 3DGS partitioning strategies fail for 360° panoramic scenes?\",\"answer\":\"Conventional partitioning depends on local visibility within camera frustums. Panoramic ERP scenes have omnipresent visibility, which removes the discriminative constraints and causes block-wise optimization to degenerate into inefficient global training.\"},{\"question\":\"What is the core idea of PanoLOG’s two-stage coarse-to-fine framework?\",\"answer\":\"PanoLOG first performs a global coarse stage to obtain reliable geometry using sky-sphere modeling and panoramic depth supervision. Then it refines reconstruction by using G2PS to build adaptive bounding volumes and to assign cameras to spatial blocks based on gradient-based importance scoring.\"},{\"question\":\"How does G2PS determine adaptive spatial boundaries and camera–block allocation?\",\"answer\":\"G2PS uses parallax-driven uncertainty to expand and stabilize reconstruction regions via depth uncertainty analysis. In the refinement stage, it evaluates each camera’s contribution to different blocks using gradient signals from the coarse stage, enabling gradient-based camera importance scoring.\"}]","Geometry and Gradient based Partitioning for Panoramic Outdoor Reconstruction | PDF",1784177725,40,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":29},"geometry-and-gradient-based-partitioning-for-panoramic-outdoor-reconstruction","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/geometry-and-gradient-based-partitioning-for-panoramic-outdoor-reconstruction/82035/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-07-29","2026-07-16",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"Why do existing 3DGS partitioning strategies fail for 360° panoramic scenes?","Question",{"text":77,"@type":78},"Conventional partitioning depends on local visibility within camera frustums. Panoramic ERP scenes have omnipresent visibility, which removes the discriminative constraints and causes block-wise optimization to degenerate into inefficient global training.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"What is the core idea of PanoLOG’s two-stage coarse-to-fine framework?",{"text":82,"@type":78},"PanoLOG first performs a global coarse stage to obtain reliable geometry using sky-sphere modeling and panoramic depth supervision. Then it refines reconstruction by using G2PS to build adaptive bounding volumes and to assign cameras to spatial blocks based on gradient-based importance scoring.",{"name":84,"@type":75,"acceptedAnswer":85},"How does G2PS determine adaptive spatial boundaries and camera–block allocation?",{"text":86,"@type":78},"G2PS uses parallax-driven uncertainty to expand and stabilize reconstruction regions via depth uncertainty analysis. In the refinement stage, it evaluates each camera’s contribution to different blocks using gradient signals from the coarse stage, enabling gradient-based camera importance scoring.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,112,116,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":108,"doc_module":4,"doc_module_name":47,"category_name":109,"show_sort_weight":110,"slug":111},5,"Comic",60,"comic",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":30,"slug":119},7,"Healthcare","healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":47,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":47,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":47,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":47,"category_name":137,"show_sort_weight":108,"slug":138},19,"General","general"]