[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-650536-105":59,"doc-detail-650536-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","residential-floor-plan-recognition-and-reconstruction","Residential Floor Plan Recognition and Reconstruction","","Recognition and reconstruction of residential floor plan drawings are critical yet challenging for design, decoration, and architectural remodeling. An automatic deep-learning framework is presented to accurately recognize room structure, type, and size, and to generate vectorized 3D reconstruction outputs. It leverages deep segmentation and detection networks, a key-point detection and cluster analysis pipeline for room scale estimation, and an iterative optimization-based vectorization method to improve accuracy and generalization, outperforming prior work, particularly for inclined wall detection.",{"@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/residential-floor-plan-recognition-and-reconstruction/650536/",{"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/residential-floor-plan-recognition-and-reconstruction/650536.png","ImageObject",300,407,{"name":92,"@type":93},"CatatanPagi","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-30","2026-10-24",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 paper address?","Question",{"text":112,"@type":113},"It addresses recognizing and reconstructing residential floor plan drawings from raster images, with an emphasis on recovering vectorized room information and real physical sizes.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does the proposed method achieve vectorized 3D reconstruction?",{"text":117,"@type":113},"It uses deep segmentation and detection networks to extract room structural information, then employs key-point detection and cluster analysis to estimate room scales, followed by an iterative optimization-based vectorization process to produce vectorized 3D results.",{"name":119,"@type":110,"acceptedAnswer":120},"What technical contributions does the system include?",{"text":121,"@type":113},"The paper contributes an end-to-end recognition and reconstruction framework, an iterative optimization vectorization method for extracting horizontal/vertical/inclined walls, a systematic scale recognition method to support real-size 3D reconstruction, and a dataset with 7000 annotated residential apartment images.","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},650536,1792919476,{"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":52,"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":143,"read_time":144},962090894170,"https://ap-avatar.wpscdn.com/davatar_6f874abed73319feea01a86fa6f0fab8","Residential 􀀃oor plan recognition and reconstruction  \nXiaolei Lv, Shengchu Zhao, Xinyang Yu, Binqiang Zhao Alibaba Group  \n{yanjun.lxl, [shengchu.sc](shengchu.sc), xinyang.yu, [binqiang.zhao](binqiang.zhao}@alibaba-inc.com)[}](binqiang.zhao}@alibaba-inc.com)[@alibaba-inc.com](binqiang.zhao}@alibaba-inc.com)  \nAbstract  \nRecognition and reconstruction of residential 􀀃oor plan drawings are important and challenging in design, decoration, and architectural remodeling 􀀂elds. An automatic framework is provided that accurately recognizes the structure, type, and size of the room, and outputs vectorized 3D reconstruction results. Deep segmentation and detection neural networks are utilized to extract room structural information. Key points detection network and cluster analysis are utilized to calculate scales of rooms. The vectorization of room information is processed through an iterative optimization-based method. The system signi􀀂cantly increases accuracy and generalization ability, compared with existing methods. It outperforms other systems in 􀀃oor plan segmentation and vectorization process, especially inclined wall detection.  \n1. Introduction  \nArchitectural 􀀃oor plans, scaled drawings of apartmentsand building spaces, are utilized to assist users to decorate rooms, design furniture layout and remodel indoor spaces [14, 22] . Floor plan images, created by speci􀀂c professional designers and architects, are rendered from vector-graphics representation, generated by AutoCAD [1], Sketchup [7] and HomeStyler [3] . However, after the above rasterization process, designers cannot modify the structure of the room and redesign 􀀃exibly. Therefore, accurately recovering vectorized information from pixel images becomes an urgent problem to be solved.  \nThe above-mentioned problems have existed for decades. The generalization ability of traditional methods is weak, the whole process is dif􀀂cult to automate, and a large amount of manual participation is required. The recognition of 􀀃oor plans needs to consider various information as shown in Figure 1, such as room structure, type, symbols, text and scale. In recent years, with the rapid development of deep learning technology, related methods have made great progress in generalization. However, it is still a challenging problem to organically integrate various infor-  \nFigure 1 . Pixel (top) and vector (bottom) 􀀃oor plan image. The pixel 􀀃oor plan image (top) contains various information, such as room structural elements (magenta), text (blue), symbols (cyan), scales (yellow) . The original image comes from HomeStyler [3] .  \nmation and perform accurate vector reconstruction of the actual physical size.  \nA novel 􀀃oor plan recognition and reconstruction system, based on deep learning, is proposed, which combines the multi-modal information of 􀀃oor plan images, such as room structure, type, symbols, text and scale. The system signi􀀂cantly outperforms existing methods, as shown in Table 1. Furthermore, we propose a dataset containing 7000 annotated images of residential apartments.  \nOur approach to residential 􀀃oor plan recognition and reconstruction includes a number of technical contributions:  \n• An automatic framework for residential 􀀃oor plan recognition and reconstruction that accurately recognizes the structure, type, and size of the room, and outputs vectorized 3D reconstruction results.  \n• An iterative optimization-based vectorization method for 􀀃oor plan images that can accurately extract horizontal, vertical and inclined walls.  \n• A systematic method for recognizing the scale of 􀀃oor plan images that can assist the 3D reconstruction of real physical size.  \n• The largest residential 􀀃oor plan image dataset with  \ndetailed vectorized and pixel annotations.  \n2. Related work  \nFloor plan drawings are crucial in real estate and design area. Systems of analyzing 􀀃oor plans have existed for decades. Traditional methods [8, 9, 10] focus on directly processing low-level features. Th","cbCaiaYS4Ao2nj2b","https://ap.wps.com/l/cbCaiaYS4Ao2nj2b","pdf",2554523,"English","# Introduction\n## Related work","[{\"question\":\"What problem does the paper address?\",\"answer\":\"It addresses recognizing and reconstructing residential floor plan drawings from raster images, with an emphasis on recovering vectorized room information and real physical sizes.\"},{\"question\":\"How does the proposed method achieve vectorized 3D reconstruction?\",\"answer\":\"It uses deep segmentation and detection networks to extract room structural information, then employs key-point detection and cluster analysis to estimate room scales, followed by an iterative optimization-based vectorization process to produce vectorized 3D results.\"},{\"question\":\"What technical contributions does the system include?\",\"answer\":\"The paper contributes an end-to-end recognition and reconstruction framework, an iterative optimization vectorization method for extracting horizontal/vertical/inclined walls, a systematic scale recognition method to support real-size 3D reconstruction, and a dataset with 7000 annotated residential apartment images.\"}]","Residential Floor Plan Recognition and Reconstruction | PDF",1792831900,25]