[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82268-en":3,"doc-seo-82268-105":29,"detail-sidebar-cat-0-en-105":91},{"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":20,"is_downloadable":20,"audit_status":20,"page_count":21,"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},82268,962075114101,"Seraphina","https://ap-avatar.wpscdn.com/avatar/e000253a75eb197efd?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780044092746381165",8,"Research & Report","Glob3R: Global Structure-from-Motion with 3D Foundation Models","Glob3R presents a global structure-from-motion (SfM) reconstruction method built on 3D geometric foundation models. While feed-forward models can directly predict camera poses and dense geometry, their outputs often contain inaccuracies and scale drift when applied to long sequences or large unordered image sets. Glob3R optimizes these geometric predictions by converting predicted image warps into sparse, reliable multi-view feature tracks for global optimization. A keyframe sliding-window association, followed by motion averaging and bundle adjustment, refines poses, improves consistency, and supports dense geometry recovery for robust neural rendering.","arXiv :2607 .09225v1 [ cs .CV] 10 Jul 2026  \nGLOB3R: GLOBAL STRUCTURE-FROM-MOTION WITH 3D FOUNDATION MODELS  \nJunyuan Deng1,2 ∗ Heng Li1 ∗ Kejie Qiu2 ∗ Lingteng Qiu2 Rui Peng2 Weichao Shen2 Weihao Yuan3 Siyu Zhu4 Zilong Dong2 Ping Tan 1†  \n1The Hong Kong University of Science and Technology 2Tongyi Lab, Alibaba Group  \n3Nanjing University 4Fudan University  \nABSTRACT  \nRecent 3D geometric foundation models, such as VGGT, provide robust feedforward 3D reconstruction by directly predicting camera poses and 3D scene points from input images. However, their results remain inaccurate, and scaling them to long sequences or large unordered image sets typically requires chunk-wise processing, which can introduce drift and inconsistency. We present Glob3R, a global SfM-style reconstruction built on 3D foundation models. Our key idea is to explicitly optimize feed-forward geometric predictions. To this end, we augment a frozen Pi3X backbone with a lightweight dense matching head that predicts image warps between selected reference frames and neighboring views. These dense warps are converted into sparse but reliable multi-view feature tracks, which provide correspondence constraints for global optimization. We further introduce a keyframe-based sliding-window association strategy that propagates tracks and relative poses across overlapping windows, enabling scalable reconstruction. Finally, we perform global motion averaging and bundle adjustment to refine camera poses, reduce scale inconsistencies, and recover dense scene geometry. Extensive experiments on indoor, outdoor, large-scale driving, and unordered SfM benchmarks demonstrate that Glob3R achieves robust and accurate reconstruction. It consistently improves over feed-forward foundation-model baselines and recent scalable reconstruction methods, while being more robust than classical SfM pipelines. The refined poses also lead to higher-quality neural rendering, validating the benefit of combining foundation-model priors with global geometric optimization. Project page: [https://junyuandeng.github.io/Glob3r/](https://junyuandeng.github.io/Glob3r/)  \n1 INTRODUCTION  \nReconstructing 3D scenes from image collections remains a fundamental challenge in computer vision, serving as a cornerstone for applications such as augmented reality (AR), robotics, autonomous navigation, and neural rendering. Recently, learning-based 3D geometric foundation models, including DUSt3R [59], VGGT [57], Pi3X [61], and other recent methods [8, 11, 16, 23, 45, 68], have emerged as a new reconstruction paradigm. Departing from traditional pipelines, these models directly predict camera poses and dense per-pixel geometry, namely depth or scene coordinates, from arbitrary image sets in a feed-forward manner. This capability offers a robust and highly efficient initialization for 3D reconstruction from both ordered image sequences and unordered image collections. Despite these advances, existing 3D geometric foundation models still suffer from limited accuracy and scalability. Their feed-forward predictions provide strong global priors, but the estimated camera poses and scales are often only approximately correct, which limits their use  \nin high-fidelity applications such as Neural Radiance Fields (NeRF) [35] and other view-synthesis pipelines. A key reason is that many models are trained with Structure from Motion (SfM)-derived  \n∗Equal contribution.†Corresponding authors. E-mail: [pingtan@ust.hk](pingtan@ust.hk)  \nsupervision, where poses and geometry are generated by tools such as COLMAP [42] rather than measured ground truth, thereby transferring the noise and bias of the reconstruction pipeline to the learned predictions. Meanwhile, GPU memory constraints make it difficult to process long sequences or large image collections in a single forward pass. Recent VGGT-based methods [10, 32, 33, 67] address this by splitting input into chunks and aligning chunk-level predictions with simple SE(3) or Sim(3) tra","cbCaigB8pqT2LVcr","https://ap.wps.com/l/cbCaigB8pqT2LVcr","pdf",18124065,1,25,"English","en",105,"# Abstract\n# Introduction","[{\"question\":\"Glob3R与现有3D几何基础模型的主要区别是什么？\",\"answer\":\"Glob3R在3D基础模型的前馈预测基础上，引入显式的几何优化流程：将由基础模型预测的图像warp转化为多视角特征轨迹，并据此进行全局优化与精化。\"},{\"question\":\"为什么仅依赖前馈预测会影响长序列或大规模无序图像集的重建精度？\",\"answer\":\"前馈预测得到的相机位姿与尺度通常只有近似正确，且受限于跨块约束不足或累计误差，容易产生漂移与尺度不一致，从而限制高精度应用。\"},{\"question\":\"Glob3R如何提升可扩展性并保持跨窗口一致性？\",\"answer\":\"Glob3R采用基于关键帧的滑动窗口关联策略，在相互重叠的窗口间传播轨迹与相对位姿。随后通过全局运动平均与束调整来进一步细化相机位姿并恢复更一致的稠密场景几何。\"}]",1784179290,63,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":27},"glob3r-global-structure-from-motion-with-3d-foundation-models","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"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":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/glob3r-global-structure-from-motion-with-3d-foundation-models/82268/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Glob3R与现有3D几何基础模型的主要区别是什么？","Question",{"text":75,"@type":76},"Glob3R在3D基础模型的前馈预测基础上，引入显式的几何优化流程：将由基础模型预测的图像warp转化为多视角特征轨迹，并据此进行全局优化与精化。","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"为什么仅依赖前馈预测会影响长序列或大规模无序图像集的重建精度？",{"text":80,"@type":76},"前馈预测得到的相机位姿与尺度通常只有近似正确，且受限于跨块约束不足或累计误差，容易产生漂移与尺度不一致，从而限制高精度应用。",{"name":82,"@type":73,"acceptedAnswer":83},"Glob3R如何提升可扩展性并保持跨窗口一致性？",{"text":84,"@type":76},"Glob3R采用基于关键帧的滑动窗口关联策略，在相互重叠的窗口间传播轨迹与相对位姿。随后通过全局运动平均与束调整来进一步细化相机位姿并恢复更一致的稠密场景几何。","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]