[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83063-en":3,"doc-seo-83063-105":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":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},83063,13056703019404,"Miles","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","APVI-SLAM Real-Time Acoustic-Pressure-Visual-Inertial Localization and Photorealistic Mapping System in Complex Underwater Environment","APVI-SLAM delivers real-time multi-sensor fusion SLAM for accurate underwater localization and photorealistic mapping in environments where visual features degrade and visual-inertial estimators diverge. The system combats intermittent visual failure using reliability-aware sensor fusion that dynamically reweights estimators and a sliding-window freezing strategy to quickly recover tracking. For high-fidelity reconstruction, a quadtree-guided mapping module enables incremental water-medium modeling and 3D Gaussian optimization. A coral-reef surveying dataset with synchronized multimodality data supports benchmark evaluation; experiments confirm state-of-the-art real-time performance.","APVI-SLAM: Real-Time Acoustic-Pressure-Visual-Inertial Localization and Photorealistic Mapping System in Complex Underwater  \nEnvironment  \nHanwen Zhang 1 , Yipeng Zhu 1 , Xiaopeng Guo 1 , Huajian Huang2 ,†, Sai-Kit Yeung3  \narXiv :2607 .06222v 1 [ cs .RO] 7 Jul 2026  \nAbstract—Extreme subsea environments often cause severe feature de-gradation and estimator divergence in underwater visual-inertial SLAM. Although sensors like Doppler Velocity Logs (DVL) and pressure gauges provide auxiliary constraints, robust multi-sensor fusion during intermittent visual failure remains challenging. To address this, we present APVI-SLAM, a real-time multi-sensor fusion SLAM system that achieves both accurate underwater localization and photorealistic mapping. Our approach introduces a reliability-aware localization framework that dynamically reweights sensor estimators and employs a sliding-window freezing strategy to recover from tracking failures, substantially enhancing system robustness. Furthermore, for high-fidelity scenes reconstruction, we propose an efficient quadtree-guided mapping module that facilitates incremental water-medium modeling and 3D Gaussian optimization. Recognizing the lack of benchmark for underwater mapping evaluation, we also contribute a coral reef surveying dataset with synchronized multi-modality data. Extensive experiments on public and our proposed benchmarks demonstrate that APVI-SLAM achieves state-of-the-art localization and reconstruction quality at real-time speeds.  \nI. INTRODUCTION  \nUnderwater Visual–Inertial Simultaneous Localization and Mapping (SLAM) plays an important role in underwater robotics. Recovering photorealistic models of underwater environments facilitates marine exploration and survey tasks.  \nHowever, rapid light attenuation, suspended particles, and ocean disturbances severely degrade imaging quality and limit the effective sensing range [1], [2], [3], often leading to unstable pose estimation and compromised reconstruction quality in Visual–Inertial (VI) SLAM systems.  \nWhile integrating complementary measurements into factor graphs improves Visual-Inertial SLAM robustness [4],[5],[6] in complex underwater environments, tightly-coupled multi-sensor estimators remain vulnerable. Since underwater visual feature loss is often persistent, visual degradation can corrupt the overall system estimation, even with covariance-based noise modeling [7] . Moreover, complete visual dropouts force a brittle re-initialization of the VI component [8], demanding time-consuming scale and covariance re-estimation. Beyond localization, existing underwater  \n† Corresponding Author  \n1Hanwen Zhang, Yipeng Zhu and Xiaopeng Guo are with the Division of Integrative Systems and Design, Hong Kong University of Science and Technology (email: {hzhangfr, yzhudg, [xguoay](xguoay}@connect.ust.hk)[}](xguoay}@connect.ust.hk)[@connect.ust.hk](xguoay}@connect.ust.hk))  \n2Huajian Huang is with the School of Interdisciplinary Science, Beijing Institute of Technology (email: [huajian@bit.edu.cn](huajian@bit.edu.cn))  \n3 Sai-Kit Yeung is with the Division of Integrative Systems and Design, the Department of Computer Science and Engineering, and the Department of Ocean Science, Hong Kong University of Science and Technology (email: [saikit@ust.hk](saikit@ust.hk))  \n\n| Physical Platform\u003Cbr>Sensor Housing Industrial\u003Cbr> |  |  | \u003Cbr>\u003Cbr>Sensor\u003Cbr>Housing\u003Cbr>\u003Cbr>\u003Cbr>Pressure\u003Cbr>\u003Cbr>WaterLinked A50 DVL |  |\n| --- | --- | --- | --- | --- |\n| Input Images |  |  |  |  |\n| Low-texture Turbid Nonstruture |  |  |  |  |\n| APVI-SLAM |  |  |  |  |\n\nFig. 1. The proposed APVI-SLAM enables real-time multi-sensor fusion for robust localization and photorealistic mapping in complex underwater environments, featuring deployment on an underwater robot and strong applicability to practical marine tasks.  \nsystems fail to reconstruct scenes with high photometric realism and structural consistency.  \nRecently, 3D Gaussian Splatting (3DGS) [9] has ","cbCaih6hjFQFVZeh","https://ap.wps.com/l/cbCaih6hjFQFVZeh","pdf",8511455,6,1,"English","en",105,"# Introduction\n## Challenges in Underwater Visual–Inertial SLAM\n## Proposed APVI-SLAM Contributions","[{\"question\":\"What problem does APVI-SLAM target in underwater visual-inertial SLAM?\",\"answer\":\"It targets severe feature degradation that causes unstable pose estimation and estimator divergence, especially during persistent or intermittent visual failure in complex underwater conditions.\"},{\"question\":\"How does APVI-SLAM improve robustness when visual tracking fails?\",\"answer\":\"It uses reliability-aware localization that dynamically reweights sensor estimators and a sliding-window freezing strategy to recover visual tracking quickly during failures.\"},{\"question\":\"How does APVI-SLAM achieve photorealistic underwater scene reconstruction?\",\"answer\":\"It introduces a quadtree-guided mapping module that supports incremental water-medium modeling and performs 3D Gaussian optimization for high-fidelity reconstruction.\"}]",1784184952,20,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":27},"apvi-slam-real-time-acoustic-pressure-visual-inertial-localization-and-photorealistic-mapping-system-in-complex-underwater-environment","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"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":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/apvi-slam-real-time-acoustic-pressure-visual-inertial-localization-and-photorealistic-mapping-system-in-complex-underwater-environment/83063/",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","2026-07-24","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},"What problem does APVI-SLAM target in underwater visual-inertial SLAM?","Question",{"text":75,"@type":76},"It targets severe feature degradation that causes unstable pose estimation and estimator divergence, especially during persistent or intermittent visual failure in complex underwater conditions.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does APVI-SLAM improve robustness when visual tracking fails?",{"text":80,"@type":76},"It uses reliability-aware localization that dynamically reweights sensor estimators and a sliding-window freezing strategy to recover visual tracking quickly during failures.",{"name":82,"@type":73,"acceptedAnswer":83},"How does APVI-SLAM achieve photorealistic underwater scene reconstruction?",{"text":84,"@type":76},"It introduces a quadtree-guided mapping module that supports incremental water-medium modeling and performs 3D Gaussian optimization for high-fidelity reconstruction.","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},[93,97,101,105,110,114,119,122,126,129,133],{"id":21,"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":20,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},"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":106,"slug":136},19,"General","general"]