{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,15]],"date-time":"2025-11-15T10:28:27Z","timestamp":1763202507405,"version":"build-2065373602"},"reference-count":65,"publisher":"Wiley","issue":"2","license":[{"start":{"date-parts":[[2021,6,4]],"date-time":"2021-06-04T00:00:00Z","timestamp":1622764800000},"content-version":"vor","delay-in-days":34,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Computer Graphics Forum"],"published-print":{"date-parts":[[2021,5]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Recent years have seen a proliferation of new digital products for the efficient management of indoor spaces, with important applications like emergency management, virtual property showcasing and interior design. While highly innovative and effective, these products rely on accurate 3D models of the environments considered, including information on both architectural and non\u2010permanent elements. These models must be created from measured data such as RGB\u2010D images or 3D point clouds, whose capture and consolidation involves lengthy data workflows. This strongly limits the rate at which 3D models can be produced, preventing the adoption of many digital services for indoor space management.<\/jats:p><jats:p>We provide a radical alternative to such data\u2010intensive procedures by presenting<jats:italic>Walk2Map<\/jats:italic>, a data\u2010driven approach to generate floor plans only from trajectories of a person walking inside the rooms. Thanks to recent advances in data\u2010driven inertial odometry, such minimalistic input data can be acquired from the IMU readings of consumer\u2010level smartphones, which allows for an effortless and scalable mapping of real\u2010world indoor spaces. Our work is based on learning the latent relation between an indoor walk trajectory and the information represented in a floor plan: interior space footprint, portals, and furniture. We distinguish between recovering area\u2010related (interior footprint, furniture) and wall\u2010related (doors) information and use two different neural architectures for the two tasks: an image\u2010based Encoder\u2010Decoder and a Graph Convolutional Network, respectively. We train our networks using scanned 3D indoor models and apply them in a cascaded fashion on an indoor walk trajectory at inference time.<\/jats:p><jats:p>We perform a qualitative and quantitative evaluation using both trajectories simulated from scanned models of interiors and measured, real\u2010world trajectories, and compare against a baseline method for image\u2010to\u2010image translation. The experiments confirm that our technique is viable and allows recovering reliable floor plans from minimal walk trajectory data.<\/jats:p>","DOI":"10.1111\/cgf.142640","type":"journal-article","created":{"date-parts":[[2021,6,4]],"date-time":"2021-06-04T16:37:32Z","timestamp":1622824652000},"page":"375-388","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":9,"title":["<i>Walk2Map<\/i>: Extracting Floor Plans from Indoor Walk Trajectories"],"prefix":"10.1111","volume":"40","author":[{"given":"Claudio","family":"Mura","sequence":"first","affiliation":[{"name":"Visualization and MultiMedia Lab University of Zurich Switzerland"},{"name":"Photogrammetry and Remote Sensing ETH Z\u00fcrich Switzerland"},{"name":"Smart Geometry Processing Group University College London"}]},{"given":"Renato","family":"Pajarola","sequence":"additional","affiliation":[{"name":"Visualization and MultiMedia Lab University of Zurich Switzerland"}]},{"given":"Konrad","family":"Schindler","sequence":"additional","affiliation":[{"name":"Photogrammetry and Remote Sensing ETH Z\u00fcrich Switzerland"}]},{"given":"Niloy","family":"Mitra","sequence":"additional","affiliation":[{"name":"Smart Geometry Processing Group University College London"},{"name":"Adobe Research London"}]}],"member":"311","published-online":{"date-parts":[[2021,6,4]]},"reference":[{"key":"e_1_2_9_2_2","doi-asserted-by":"crossref","unstructured":"BaoS. Y. FurlanA. Fei-FeiL. SavareseS.: Understanding the 3D layout of a cluttered room from multiple images. InWinter Conference on Applications of Computer Vision (WACV)(2014) pp.690\u2013697. 2","DOI":"10.1109\/WACV.2014.6836035"},{"key":"e_1_2_9_3_2","unstructured":"BengioY. LouradourJ. CollobertR. WestonJ.: Curriculum learning. InProceedings International Conference on Machine Learning (ICML)(2009) pp.41\u201348. 5"},{"key":"e_1_2_9_4_2","doi-asserted-by":"crossref","unstructured":"ChangA. DaiA. FunkhouserT. HalberM. NiessnerM. SavvaM. SongS. ZengA. ZhangY.: Matterport3D: Learning from RGB-D data in indoor environments. InProceedings International Conference on 3D Vision (3DV)(2017) pp.667\u2013676. 5","DOI":"10.1109\/3DV.2017.00081"},{"key":"e_1_2_9_5_2","doi-asserted-by":"crossref","unstructured":"CabralR. FurukawaY.: Piecewise planar and compact floorplan reconstruction from images. InProceedings Computer Vision and Pattern Recognition (CVPR)(2014) pp.628\u2013635. 2 3","DOI":"10.1109\/CVPR.2014.546"},{"key":"e_1_2_9_6_2","unstructured":"ChenC. LuX. WahlstromJ. MarkhamA. TrigoniN.: Deep neural network based inertial odometry using low-cost inertial measurement units.IEEE Transactions on Mobile Computing(2019). 2 3"},{"key":"e_1_2_9_7_2","doi-asserted-by":"crossref","unstructured":"ChenJ. LiuC. WuJ. FurukawaY.: Floor-SP: Inverse CAD for floorplans by sequential room-wise shortest path. InProceedings International Conference on Computer Vision (ICCV)(2019) pp.2661\u20132670. 3","DOI":"10.1109\/ICCV.2019.00275"},{"key":"e_1_2_9_8_2","unstructured":"ChenC. ZhaoP. LuC. X. WangW. MarkhamA. TrigoniN.: Deep learning based pedestrian inertial navigation: Methods dataset and on-device inference.arXiv preprint arXiv:2001.04061(2020). 3"},{"key":"e_1_2_9_9_2","unstructured":"FurukawaY. CurlessB. SeitzS. M. SzeliskiR.: Reconstructing building interiors from images. InProceedings International Conference on Computer Vision (ICCV)(2009) pp.80\u201387. 3"},{"key":"e_1_2_9_10_2","doi-asserted-by":"publisher","DOI":"10.1145\/3130800.3130878"},{"key":"e_1_2_9_11_2","unstructured":"FeyM. LenssenJ. E.: Fast graph representation learning with pytorch geometric.arXiv preprint arXiv:1903.02428(2019). 7"},{"key":"e_1_2_9_12_2","doi-asserted-by":"publisher","DOI":"10.1145\/2366145.2366154"},{"key":"e_1_2_9_13_2","volume":"01","author":"Gibson J.","year":"1979","journal-title":"Ecological Approach to Visual Perception"},{"key":"e_1_2_9_14_2","unstructured":"Gurobi OptimizationL.:Gurobi optimizer reference manual.www.gurobi.com 2020. [Accessed: 2020-10-05]. 7"},{"key":"e_1_2_9_15_2","unstructured":"Google:Google Indoor Maps.www.google.com\/maps\/about\/partners\/indoormaps 2020. [Accessed: 2020-10-05]. 1"},{"key":"e_1_2_9_16_2","doi-asserted-by":"crossref","unstructured":"HedauV. HoiemD. ForsythD.: Recovering the spatial layout of cluttered rooms. InProceedings International Conference on Computer Vision (ICCV)(2009) pp.1849\u20131856. 3","DOI":"10.1109\/ICCV.2009.5459411"},{"issue":"3","key":"e_1_2_9_17_2","first-page":"52","article-title":"Representation learning on graphs: Methods and applications","volume":"40","author":"Hamilton W. L.","year":"2017","journal-title":"IEEE Data Engineering Bulletin"},{"key":"e_1_2_9_18_2","unstructured":"IkehataS. YangH. FurukawaY.: Structured indoor modeling. InProceedings International Conference on Computer Vision (ICCV)(2015) pp.1323\u20131331. 2 3"},{"key":"e_1_2_9_19_2","unstructured":"IsolaP. ZhuJ.-Y. ZhouT. EfrosA. A.: Image-to-image translation with conditional adversarial networks. InProceedings Computer Vision and Pattern Recognition (CVPR)(2017) pp.1125\u20131134. 5 7 11"},{"key":"e_1_2_9_20_2","unstructured":"KingmaD. P. BaJ.: Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980(2014). 6"},{"key":"e_1_2_9_21_2","doi-asserted-by":"crossref","unstructured":"KirsanovP. GaskarovA. KonokhovF. SofiiukK. VorontsovaA. SlinkoI. ZhukovD. BykovS. BarinovaO. KonushinA.: Discoman: Dataset of indoor scenes for odometry mapping and navigation.arXiv preprint arXiv:1909.12146(2019). 5 7","DOI":"10.1109\/IROS40897.2019.8967921"},{"key":"e_1_2_9_22_2","doi-asserted-by":"crossref","unstructured":"KiyotaY.: Promoting open innovations in real estate tech: Provision of the lifull home's data set and collaborative studies. InInternational Conference on Multimedia Retrieval (ICMR) Industrial Talks(2018) pp.6\u20136. 5","DOI":"10.1145\/3206025.3210494"},{"key":"e_1_2_9_23_2","unstructured":"KipfT. N. WellingM.: Semi-supervised classification with graph convolutional networks. InInternational Conference on Learning Representations (ICLR)(2017). 6"},{"key":"e_1_2_9_24_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2004.1262177"},{"key":"e_1_2_9_25_2","doi-asserted-by":"crossref","unstructured":"LeeD. C. HebertM. KanadeT.: Geometric reasoning for single image structure recovery. InProceedings Computer Vision and Pattern Recognition (CVPR)(2009) pp.2136\u20132143. 3","DOI":"10.1109\/CVPR.2009.5206872"},{"volume-title":"Markov Random Field Modeling in Computer Vision","year":"2012","author":"Li S. Z.","key":"e_1_2_9_26_2"},{"key":"e_1_2_9_27_2","doi-asserted-by":"publisher","DOI":"10.1177\/0278364914554813"},{"key":"e_1_2_9_28_2","doi-asserted-by":"publisher","DOI":"10.1145\/3303766"},{"key":"e_1_2_9_29_2","doi-asserted-by":"crossref","unstructured":"LiuC. WuJ. FurukawaY.: FloorNet: A unified framework for floorplan reconstruction from 3D scans. InProceedings of the European Conference on Computer Vision (ECCV)(2018) pp.203\u2013219. 2 3","DOI":"10.1007\/978-3-030-01231-1_13"},{"key":"e_1_2_9_30_2","doi-asserted-by":"crossref","unstructured":"LiuC. WuJ. KohliP. FurukawaY.: Raster-to-vector: Revisiting floorplan transformation. InProceedings International Conference on Computer Vision (ICCV)(2017) pp.2195\u20132203. 4 5","DOI":"10.1109\/ICCV.2017.241"},{"key":"e_1_2_9_31_2","unstructured":"Matterport:Matterport.matterport.com 2020. [Accessed: 2020-10-05]. 1"},{"key":"e_1_2_9_32_2","doi-asserted-by":"crossref","unstructured":"Marder-EppsteinE.: Project tango. InProceedings SIGGRAPH Real-Time Live!(2016). 5","DOI":"10.1145\/2933540.2933550"},{"key":"e_1_2_9_33_2","doi-asserted-by":"publisher","DOI":"10.1145\/3306346.3322961"},{"key":"e_1_2_9_34_2","doi-asserted-by":"publisher","DOI":"10.1145\/2980179.2980223"},{"key":"e_1_2_9_35_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cag.2014.07.005"},{"key":"e_1_2_9_36_2","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.13015"},{"key":"e_1_2_9_37_2","doi-asserted-by":"publisher","DOI":"10.1145\/2010324.1964982"},{"key":"e_1_2_9_38_2","doi-asserted-by":"crossref","unstructured":"NieY. HanX. GuoS. ZhengY. ChangJ. ZhangJ. J.: Total 3D understanding: Joint layout object pose and mesh reconstruction for indoor scenes from a single image. InProceedings Computer Vision and Pattern Recognition (CVPR)(2020) pp.55\u201364. 3","DOI":"10.1109\/CVPR42600.2020.00013"},{"key":"e_1_2_9_39_2","doi-asserted-by":"crossref","unstructured":"Nist\u00e9rD. NaroditskyO. BergenJ.: Visual odometry. InProceedings Computer Vision and Pattern Recognition (CVPR)(2004) vol. 1 pp.I\u2013I. 3","DOI":"10.1109\/CVPR.2004.1315094"},{"key":"e_1_2_9_40_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.isprsjprs.2014.02.004"},{"key":"e_1_2_9_41_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.isprsjprs.2019.03.017"},{"key":"e_1_2_9_42_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cag.2015.07.008"},{"key":"e_1_2_9_43_2","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.13842"},{"key":"e_1_2_9_44_2","unstructured":"PaszkeA. GrossS. MassaF. LererA. BradburyJ. ChananG. KilleenT. LinZ. GimelsheinN. AntigaL. et al.: Pytorch: An imperative style high-performance deep learning library. InAdvances in Neural Information Processing Systems(2019) pp.8024\u20138035. 6"},{"key":"e_1_2_9_45_2","doi-asserted-by":"publisher","DOI":"10.1007\/s41095-018-0125-9"},{"key":"e_1_2_9_46_2","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.14021"},{"key":"e_1_2_9_47_2","doi-asserted-by":"crossref","unstructured":"QiS. ZhuY. HuangS. JiangC. ZhuS.-C.: Human-centric indoor scene synthesis using stochastic grammar. InProceedings Computer Vision and Pattern Recognition (CVPR)(2018) pp.5899\u20135908. 4","DOI":"10.1109\/CVPR.2018.00618"},{"key":"e_1_2_9_48_2","doi-asserted-by":"crossref","unstructured":"RonnebergerO. FischerP. BroxT.: U-net: Convolutional networks for biomedical image segmentation. InProceedings International Conference on Medical Image Computing and Computer-Assisted Intervention(2015) pp.234\u2013241. 5 6","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_9_49_2","doi-asserted-by":"crossref","unstructured":"RitchieD. WangK. LinY.-a.: Fast and flexible indoor scene synthesis via deep convolutional generative models. InProceedings Computer Vision and Pattern Recognition(2019) pp.6182\u20136190. 4","DOI":"10.1109\/CVPR.2019.00634"},{"key":"e_1_2_9_50_2","doi-asserted-by":"publisher","DOI":"10.1145\/2897824.2925867"},{"key":"e_1_2_9_51_2","doi-asserted-by":"publisher","DOI":"10.1109\/MRA.2011.943233"},{"key":"e_1_2_9_52_2","doi-asserted-by":"crossref","unstructured":"SunC. HsiaoC.-W. SunM. ChenH.-T.: Horizonnet: Learning room layout with 1D representation and pano stretch data augmentation. InProceedings Computer Vision and Pattern Recognition (CVPR)(2019) pp.1047\u20131056. 3","DOI":"10.1109\/CVPR.2019.00114"},{"key":"e_1_2_9_53_2","doi-asserted-by":"crossref","unstructured":"SimonovskyM. KomodakisN.: Dynamic edge-conditioned filters in convolutional neural networks on graphs. InProceedings Computer Vision and Pattern Recognition (CVPR)(2017) pp.3693\u20133702. 6 7","DOI":"10.1109\/CVPR.2017.11"},{"key":"e_1_2_9_54_2","unstructured":"Trimble: SketchUp.www.sketchup.com 2020. [Accessed: 2020-10-05]. 2"},{"key":"e_1_2_9_55_2","doi-asserted-by":"crossref","unstructured":"TurnerE. ZakhorA.: Watertight as-built architectural floor plans generated from laser range data. InProceedings 3DIMPVT(2012) pp.316\u2013323. 2","DOI":"10.1109\/3DIMPVT.2012.80"},{"key":"e_1_2_9_56_2","doi-asserted-by":"publisher","DOI":"10.1145\/3306346.3322941"},{"key":"e_1_2_9_57_2","doi-asserted-by":"publisher","DOI":"10.1145\/3450626.3459787"},{"key":"e_1_2_9_58_2","doi-asserted-by":"crossref","unstructured":"XuJ. StengerB. KerolaT. TungT.: Pano2cad: Room layout from a single panorama image. InProceedings Winter Conference on Applications of Computer Vision (WACV)(2017) pp.354\u2013362. 3","DOI":"10.1109\/WACV.2017.46"},{"key":"e_1_2_9_59_2","doi-asserted-by":"crossref","unstructured":"YanH. HerathS. FurukawaY.: Ronin: Robust neural inertial navigation in the wild: Benchmark evaluations and new methods.arXiv preprint arXiv:1905.12853(2019). 2 3","DOI":"10.1109\/ICRA40945.2020.9196860"},{"key":"e_1_2_9_60_2","unstructured":"YanH. ShanQ. FurukawaY.: RIDI: Robust IMU double integration. InProceedings of the European Conference on Computer Vision (ECCV)(2018) pp.621\u2013636. 2 3 5"},{"key":"e_1_2_9_61_2","doi-asserted-by":"crossref","unstructured":"YangS.-T. WangF.-E. PengC.-H. WonkaP. SunM. ChuH.-K.: Dula-net: A dual-projection network for estimating room layouts from a single RGB panorama. InProceedings Computer Vision and Pattern Recognition (CVPR)(2019) pp.3363\u20133372. 3","DOI":"10.1109\/CVPR.2019.00348"},{"key":"e_1_2_9_62_2","doi-asserted-by":"publisher","DOI":"10.1145\/2010324.1964981"},{"key":"e_1_2_9_63_2","doi-asserted-by":"publisher","DOI":"10.1145\/2185520.2185552"},{"key":"e_1_2_9_64_2","doi-asserted-by":"crossref","unstructured":"YangH. ZhangH.: Efficient 3D room shape recovery from a single panorama. InProceedings Computer Vision and Pattern Recognition (CVPR)(2016) pp.5422\u20135430. 3","DOI":"10.1109\/CVPR.2016.585"},{"key":"e_1_2_9_65_2","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.13386"},{"key":"e_1_2_9_66_2","doi-asserted-by":"crossref","unstructured":"ZhangY. SongS. TanP. XiaoJ.: Panocontext: A whole-room 3D context model for panoramic scene understanding. InProceedings European Conference on Computer Vision (ECCV)(2014) pp.668\u2013686. 3","DOI":"10.1007\/978-3-319-10599-4_43"}],"container-title":["Computer Graphics Forum"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/cgf.142640","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1111\/cgf.142640","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/cgf.142640","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,1]],"date-time":"2024-09-01T07:20:59Z","timestamp":1725175259000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/cgf.142640"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,5]]},"references-count":65,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2021,5]]}},"alternative-id":["10.1111\/cgf.142640"],"URL":"https:\/\/doi.org\/10.1111\/cgf.142640","archive":["Portico"],"relation":{},"ISSN":["0167-7055","1467-8659"],"issn-type":[{"type":"print","value":"0167-7055"},{"type":"electronic","value":"1467-8659"}],"subject":[],"published":{"date-parts":[[2021,5]]},"assertion":[{"value":"2021-06-04","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}