000106609 001__ 106609 000106609 005__ 20210902121837.0 000106609 0247_ $$2doi$$a10.1109/CVPR42600.2020.00270 000106609 0248_ $$2sideral$$a121225 000106609 037__ $$aART-2020-121225 000106609 041__ $$aeng 000106609 100__ $$aWarburg, F. 000106609 245__ $$aMapillary street-level sequences: A dataset for lifelong place recognition 000106609 260__ $$c2020 000106609 5060_ $$aAccess copy available to the general public$$fUnrestricted 000106609 5203_ $$aLifelong place recognition is an essential and challenging task in computer vision with vast applications in robust localization and efficient large-scale 3D reconstruction. Progress is currently hindered by a lack of large, diverse, publicly available datasets. We contribute with Mapillary Street-Level Sequences (SLS), a large dataset for urban and suburban place recognition from image sequences. It contains more than 1.6 million images curated from the Mapillary collaborative mapping platform. The dataset is orders of magnitude larger than current data sources, and is designed to reflect the diversities of true lifelong learning. It features images from 30 major cities across six continents, hundreds of distinct cameras, and substantially different viewpoints and capture times, spanning all seasons over a nine year period. All images are geo-located with GPS and compass, and feature high-level attributes such as road type. We propose a set of benchmark tasks designed to push state-of-the-art performance and provide baseline studies. We show that current state-of-the-art methods still have a long way to go, and that the lack of diversity in existing datasets have prevented generalization to new environments. The dataset and benchmarks are available for academic research. 000106609 536__ $$9info:eu-repo/grantAgreement/ES/DGA-FSE/T45-17R$$9info:eu-repo/grantAgreement/EC/H2020/757360/EU/Measuring with no tape/NoTape$$9This project has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No H2020 757360-NoTape$$9info:eu-repo/grantAgreement/ES/MCIU-AEI-FEDER/PGC2018-096367-B-I00 000106609 540__ $$9info:eu-repo/semantics/openAccess$$aAll rights reserved$$uhttp://www.europeana.eu/rights/rr-f/ 000106609 592__ $$a4.658$$b2020 000106609 593__ $$aSoftware$$c2020 000106609 593__ $$aComputer Vision and Pattern Recognition$$c2020 000106609 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/acceptedVersion 000106609 700__ $$aHauberg, S. 000106609 700__ $$aLópez-Antequera, M. 000106609 700__ $$aGargallo, P. 000106609 700__ $$aKuang, Y. 000106609 700__ $$0(orcid)0000-0003-1368-1151$$aCivera, J.$$uUniversidad de Zaragoza 000106609 7102_ $$15007$$2520$$aUniversidad de Zaragoza$$bDpto. Informát.Ingenie.Sistms.$$cÁrea Ingen.Sistemas y Automát. 000106609 773__ $$g(2020), 2623-2632$$pProc.- IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit.$$tProceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition$$x1063-6919 000106609 8564_ $$s4367058$$uhttps://zaguan.unizar.es/record/106609/files/texto_completo.pdf$$yPostprint 000106609 8564_ $$s2638144$$uhttps://zaguan.unizar.es/record/106609/files/texto_completo.jpg?subformat=icon$$xicon$$yPostprint 000106609 909CO $$ooai:zaguan.unizar.es:106609$$particulos$$pdriver 000106609 951__ $$a2021-09-02-10:21:11 000106609 980__ $$aARTICLE