[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86540-en":3,"doc-seo-86540-105":29,"detail-sidebar-cat-0-en-105":90},{"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":11,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},86540,962075114101,"Seraphina","https://ap-avatar.wpscdn.com/avatar/e000253a75eb197efd?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780044092746381165",8,"Research & Report","GeoGS-SLAM Online Monocular Reconstruction Using Gaussian Splatting with Geometric Priors","SLAM systems based on 3D Gaussian Splatting (3DGS) deliver strong tracking and mapping, yet often need extra geometry from depth sensors, and prior approaches using geometric priors may ignore original RGB during optimization, hurting reconstruction fidelity. GeoGS-SLAM is an online monocular dense reconstruction system that fuses a 3DGS map representation with learned geometric priors. Uncalibrated RGB is used to predict camera and scene priors, sample Gaussian primitives from both RGB and priors, then jointly optimize poses and map via coarse-to-fine photometric and geometric losses, with loop closure and pose-graph optimization.","GeoGS-SLAM: Online Monocular Reconstruction Using Gaussian  \nSplatting with Geometric Priors  \narXiv :2607 . 11184v1 [ cs .RO] 13 Jul 2026  \nWaymo TUM RGB-D Replica  \nRuilan Gao 1 , Letian Jin 1 , Yu Zhang 1 ,2 ,˚  \nMonoGS Photo-SLAM S3PO-GS Ours Ground Truth  \nFig. 1: Rendering results on three datasets. Our method produces high-fidelity reconstructions on both indoor and outdoor benchmarks, outperforming state-of-the-art monocular 3DGS-based SLAM methods.  \nAbstract—SLAM methods based on 3D Gaussian Splatting (3DGS) have demonstrated impressive tracking and mapping performance, but typically require additional geometric information from external depth sensors. Meanwhile, recent SLAM systems that leverage geometric priors from pre-trained feedforward models enable real-time dense reconstruction, yet often discard original RGB information during optimization, thus degrading overall reconstruction quality. We present GeoGSSLAM, an online monocular dense reconstruction system that combines the 3DGS-based map representation with learned geometric priors. Given uncalibrated RGB input, we first employ a feed-forward visual geometry model to predict camera and scene priors. The Gaussian scene map is then expanded by directly sampling Gaussian primitives from both RGB input and geometric priors. Camera poses and the scene map are jointly optimized through a coarse-to-fine strategy that minimizes both photometric and geometric losses. To ensure global consistency, we further incorporate online loop closure detection and pose graph optimization. Extensive experiments across indoor and outdoor benchmarks demonstrate that GeoGS-SLAM achieves superior rendering quality and tracking accuracy compared to state-of-the-art methods while maintaining online real-time performance. Project page: [https://rlgao.github.io/](https://rlgao.github.io/)[ ](https://rlgao.github.io/)geogs_slam.  \nThis work was supported by the National Natural Science Foundation of China (Grant No. 62576311), in part by NSFC 62088101 Autonomous Intelligent Unmanned Systems, and in part by Zhejiang Provincial Natural Science Foundation of China under Grant No. LD24F030001 .  \n1 State Key Laboratory of Industrial Control Technology, College of Control Science and Engineering, Zhejiang University, Hangzhou, China, 310027.  \n2 Key Laboratory of Collaborative Sensing and Autonomous Unmanned Systems of Zhejiang Province, Hangzhou, China, 310027 .  \n˚ Corresponding author: Yu Zhang (Email: [zhangyu80@zju.edu.cn](zhangyu80@zju.edu.cn)) .  \nI. INTRODUCTION  \nSimultaneous localization and mapping (SLAM) is a core problem in computer vision, serving as the foundation for applications ranging from robotics and autonomous driving to augmented reality systems. Recent advances in SLAM have been propelled by two complementary breakthroughsin view synthesis and 3D reconstruction: radiance field rendering [1] and feed-forward scene reconstruction [2] .  \nThe advent of neural radiance field (NeRF) [3] and 3D Gaussian Splatting (3DGS) [4] has fundamentally transformed scene representations in SLAM systems. In particular, 3DGS employs differentiable rasterization of 3DGaussians to achieve efficient, photorealistic rendering, and has been shown to support high-quality tracking and mapping in SLAM [5], [6] . However, existing 3DGS-based SLAM systems predominantly rely on external geometric measurements from depth sensors and exhibit degraded performance when constrained to RGB-only input [7] .  \nMore recently, feed-forward models such as DUSt3R [8] and VGGT [9] have revolutionized 3D scene reconstruction through Transformer-based architectures trained at scale. A growing number of SLAM systems now leverage geometric priors from these powerful models to achieve real-time pose estimation and dense scene reconstruction [10], [11] . However, these methods typically treat the learned priorsas the primary optimization signal while excluding original RGB observations from the optimization loop. For e","cbCaiactNcV6l7t9","https://ap.wps.com/l/cbCaiactNcV6l7t9","pdf",3643810,3,1,"English","en",105,"# Introduction\n# Method Overview\n## Geometric Prior Prediction\n## Gaussian Primitive Sampling and Joint Optimization\n## Loop Closure and Pose Graph Optimization\n# Experiments and Results","[{\"question\":\"What problem does GeoGS-SLAM address compared with prior 3DGS-based SLAM methods?\",\"answer\":\"GeoGS-SLAM targets RGB-only monocular reconstruction without relying on external depth sensors, and it avoids degrading quality caused by excluding original RGB evidence during optimization.\"},{\"question\":\"How does GeoGS-SLAM generate geometric priors from uncalibrated RGB input?\",\"answer\":\"It uses a pre-trained feed-forward visual geometry model to predict camera intrinsics/extrinsics, depth maps, and point maps, then applies scale alignment for robust multi-view guidance.\"},{\"question\":\"What optimization strategy ensures accurate and consistent reconstruction in GeoGS-SLAM?\",\"answer\":\"It jointly refines camera poses and the 3D Gaussian scene map using coarse-to-fine rendering-based optimization that minimizes photometric and geometric losses, and it adds online loop closure with pose-graph optimization for global consistency.\"}]",1784212503,20,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"geogs-slam-online-monocular-reconstruction-using-gaussian-splatting-with-geometric-priors","",{"@graph":35,"@context":84},[36,52,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,49],{"item":40,"name":41,"@type":42,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":20},"https://docshare.wps.com/document/research-report/",{"item":50,"name":13,"@type":42,"position":51},"https://docshare.wps.com/document/geogs-slam-online-monocular-reconstruction-using-gaussian-splatting-with-geometric-priors/86540/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":23,"description":14,"dateModified":60,"datePublished":61,"encodingFormat":59,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":40,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":20},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What problem does GeoGS-SLAM address compared with prior 3DGS-based SLAM methods?","Question",{"text":74,"@type":75},"GeoGS-SLAM targets RGB-only monocular reconstruction without relying on external depth sensors, and it avoids degrading quality caused by excluding original RGB evidence during optimization.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does GeoGS-SLAM generate geometric priors from uncalibrated RGB input?",{"text":79,"@type":75},"It uses a pre-trained feed-forward visual geometry model to predict camera intrinsics/extrinsics, depth maps, and point maps, then applies scale alignment for robust multi-view guidance.",{"name":81,"@type":72,"acceptedAnswer":82},"What optimization strategy ensures accurate and consistent reconstruction in GeoGS-SLAM?",{"text":83,"@type":75},"It jointly refines camera poses and the 3D Gaussian scene map using coarse-to-fine rendering-based optimization that minimizes photometric and geometric losses, and it adds online loop closure with pose-graph optimization for global consistency.","https://schema.org",{"og:url":50,"og:type":86,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":88,"canonical":50},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":21,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":28,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":28,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":28,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]