[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-663216-105":59,"doc-detail-663216-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","bim-module-for-deep-learning-driven-parametric-ifc-reconstruction","BIM Module for Deep Learning-driven Parametric IFC Reconstruction","","Building Information Models (BIM) increasingly rely on advanced software, plug-ins, and APIs, yet geometric modelling is often limited to interactive work and proprietary platforms, slowing flexible and efficient reconstruction. This paper presents a DL-driven BIM module for parametric IFC reconstruction for the Scan-to-BIM Challenge at the CVPR 2024 Workshop, where it achieved second place. The pipeline reconstructs primary and secondary elements from unstructured TLS point clouds using CNN and transformer-based methods with late-fusion instance segmentation. It outputs IFC elements, improving modelling accuracy, parameter estimation, and downstream consistency, with strong results across multiple datasets that demonstrate the value of deep learning.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/bim-module-for-deep-learning-driven-parametric-ifc-reconstruction/663216/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/bim-module-for-deep-learning-driven-parametric-ifc-reconstruction/663216.png","ImageObject",300,407,{"name":92,"@type":93},"Kurz","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-29","2026-10-28",true,{"@type":102,"interactionType":103,"userInteractionCount":19},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does the proposed DL-driven BIM module address?","Question",{"text":112,"@type":113},"It targets the limited flexibility and efficiency of geometric modelling in Scan-to-BIM workflows by providing a largely automated reconstruction pipeline from unstructured point clouds.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does the pipeline reconstruct building elements for BIM?",{"text":117,"@type":113},"It uses deep learning with CNNs and transformer networks plus late-fusion instance segmentation across 2D and 3D modalities to identify class-specific elements before generating IFC outputs.",{"name":119,"@type":110,"acceptedAnswer":120},"What inputs and output formats does the method use?",{"text":121,"@type":113},"The method starts from unstructured point cloud data captured via Terrestrial Laser Scanning (TLS) and ultimately produces Industry Foundation Classes (IFC) elements for the BIM framework.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},663216,1793225827,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":19,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":39,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":129,"read_time":46},2336478945635,"https://ap-avatar.wpscdn.com/davatar_6f874abed73319feea01a86fa6f0fab8","BIM Module for Deep Learning-driven parametric IFC reconstruction  \nOscar Roman 1,2, Maarten Bassier 3, Sam De Geyter 3, Heinder De Winter3, Elisa Mariarosaria Farella 2, Fabio Remondino 2  \n1 Department Information Engineering and Computer Science (IECS), University of Trento, Trento, Italy-oscar.roman@unitn.it  \n2 3D Optical Metrology (3DOM) unit, Bruno Kessler Foundation (FBK), Trento, Italy  \nweb: [http://3dom.fbk.eu](http://3dom.fbk.eu) – email: [oroman@fbk.eu](oroman@fbk.eu), [elifarella@fbk.eu](elifarella@fbk.eu), [remondino@fbk.eu](remondino@fbk.eu)  \n3 Dept. of Civil Engineering, TC Construction-Geomatics, KU Leuven-Faculty of Engineering Technology, Ghent, Belgium [maarten.bassier@kuleuven.be](maarten.bassier@kuleuven.be), [sam.degeyter@kuleuven.be](sam.degeyter@kuleuven.be), [heinder.dewinter@kuleuven.be](heinder.dewinter@kuleuven.be)  \nKeywords: Automation in constructions, Deep-Learning, Scan-to-BIM, Semantic segmentation, Computational Geometry  \nAbstract  \nThe creation of Building Information Models (BIM) is driven by cutting-edge software applications, plug-ins, and APIs that constitute the backbone of BIM authoring tools. While free tools and APIs offer visualization and customization options, geometric modelling remains largely restricted to interactive work and proprietary platforms, which sometimes limits flexibility and efficiency. There are still only a few comprehensive workflows that fully automate the reconstruction of building elements from reality-based surveyed data. This paper introduces an innovative reconstruction pipeline developed for the Scan-to-BIM Challenge at the CVPR 2024 Workshop, where it achieved second place in the competition. A deep learning (DL)-driven BIM Module for parametric IFC reconstruction is designed to accurately reconstruct both primary and secondary building elements within a BIM framework, starting from unstructured point cloud data captured via Terrestrial Laser Scanning (TLS) . By leveraging DL techniques, particularly Convolutional Neural Networks (CNNs) and Transformers Networks (PTv3), our approach uses late fusion instance segmentation across both 2D and 3D modalities to accurately identify and reconstruct class-specific elements. The pipeline ultimately generates Industry Foundation Classes (IFC) elements, enhancing modelling accuracy, parameter estimation, and consistency in subsequent stages. Results highlight the pipeline’s strong performance on various datasets, underscoring the crucial role of DL in advancing Scan-to-BIM workflows.  \n1. Introduction  \nThe Scan-to-BIM concept has gained significant attention due to its increasing multi-sector application and interoperability, leading to the development of various BIM reconstruction methods from point cloud data, each utilizing different strategies (Bassier et al., 2016; Rashdi et al., 2022) . Traditionally, the BIM industry has been dominated by proprietary authoring tools, which can sometimes hinder workflow efficiency for companies and engineering studios. As the demand for BIM solutions continues to rise, especially in Architecture, Engineering, and Construction (AEC) sector (Rocha and Mateus, 2021), there is a growing need for more efficient and open-source alternatives able to deliver comparable functionality and efficiency. Additionally, the growing focus on automating Scan-to-BIM processes has led to more efficient and advanced methods for reconstructing models directly from point cloud data.  \nDespite the AEC industry's growing demand for Scan-to-BIM procedures, challenges in automation persist. Key benchmarks on this topic, such as ISPRS and CVPR 2024, focus on complex indoor scenes using single-storey LiDAR or RGBD data, with strict requirements for Industry Foundation Classes (IFC) and evaluations based on completeness, correctness, and accuracy. While comprehensive BIM benchmarks remain limited, advancements in machine learning (ML) and deep learning (DL) have significantly enhanced automation in S","cbCaiddFSMjV2zSS","https://ap.wps.com/l/cbCaiddFSMjV2zSS","pdf",15375676,"English","# Introduction\n## Objectives\n# Background and related works\n## Segmentation and classification","[{\"question\":\"What problem does the proposed DL-driven BIM module address?\",\"answer\":\"It targets the limited flexibility and efficiency of geometric modelling in Scan-to-BIM workflows by providing a largely automated reconstruction pipeline from unstructured point clouds.\"},{\"question\":\"How does the pipeline reconstruct building elements for BIM?\",\"answer\":\"It uses deep learning with CNNs and transformer networks plus late-fusion instance segmentation across 2D and 3D modalities to identify class-specific elements before generating IFC outputs.\"},{\"question\":\"What inputs and output formats does the method use?\",\"answer\":\"The method starts from unstructured point cloud data captured via Terrestrial Laser Scanning (TLS) and ultimately produces Industry Foundation Classes (IFC) elements for the BIM framework.\"}]","BIM Module for Deep Learning-driven Parametric IFC Reconstruction | PDF"]