{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,8,2]],"date-time":"2025-08-02T16:49:56Z","timestamp":1754153396200,"version":"3.41.2"},"reference-count":55,"publisher":"Cambridge University Press (CUP)","issue":"5","license":[{"start":{"date-parts":[[2025,3,31]],"date-time":"2025-03-31T00:00:00Z","timestamp":1743379200000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/www.cambridge.org\/core\/terms"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Robotica"],"published-print":{"date-parts":[[2025,5]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Among various deep learning-based SLAM systems, many exhibit low accuracy and inadequate generalization on non-training datasets. The deficiency in generalization ability can result in significant errors within SLAM systems during real-world applications, particularly in environments that diverge markedly from those represented in the training set. This paper presents a methodology to enhance the generalization capabilities of deep learning SLAM systems. It leverages their superior performance in feature extraction and introduces Exponential Moving Average (EMA) and Bayes online learning to improve generalization and localization accuracy in varied scenarios. Experimental validation, utilizing Absolute Trajectory Error (ATE) metrics on the dataset, has been conducted. The results demonstrate that this method effectively reduces errors by <jats:inline-formula><jats:alternatives><jats:inline-graphic xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" mime-subtype=\"png\" xlink:href=\"S0263574725000360_inline1.png\"\/><jats:tex-math>\n$20\\%$\n<\/jats:tex-math><\/jats:alternatives><\/jats:inline-formula> on the EuRoC dataset and by <jats:inline-formula><jats:alternatives><jats:inline-graphic xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" mime-subtype=\"png\" xlink:href=\"S0263574725000360_inline2.png\"\/><jats:tex-math>\n$35\\%$\n<\/jats:tex-math><\/jats:alternatives><\/jats:inline-formula> on the TUM dataset, respectively.<\/jats:p>","DOI":"10.1017\/s0263574725000360","type":"journal-article","created":{"date-parts":[[2025,3,31]],"date-time":"2025-03-31T05:49:10Z","timestamp":1743400150000},"page":"1660-1674","source":"Crossref","is-referenced-by-count":0,"title":["Improve generalization for neural visual-SLAM with Bayes online learning"],"prefix":"10.1017","volume":"43","author":[{"given":"Jun","family":"Liu","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-4996-1027","authenticated-orcid":false,"given":"Haihang","family":"Deng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"56","published-online":{"date-parts":[[2025,3,31]]},"reference":[{"key":"S0263574725000360_ref5","doi-asserted-by":"publisher","DOI":"10.1177\/0278364915620033"},{"key":"S0263574725000360_ref21","doi-asserted-by":"crossref","first-page":"2100","DOI":"10.1109\/IROS.2013.6696650","volume-title":"2013 IEEE\/RSJ International Conference on Intelligent Robots and Systems","author":"Kerl","year":"2013"},{"key":"S0263574725000360_ref37","unstructured":"[37] Tang, C. and Tan, P. , \u201cBA-Net: Dense bundle adjustment network,\u201d (2018) arXiv preprint arXiv: 1806.04807"},{"key":"S0263574725000360_ref24","first-page":"6816","article-title":"Detecting and adapting to irregular distribution shifts in Bayesian online learning","volume":"34","author":"Li","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"S0263574725000360_ref39","unstructured":"[39] Teed, Z. and Deng, J. , \u201cDeepv2d: Video to depth with differentiable structure from motion,\u201d (2018) arXiv preprint arXiv: 1812.04605"},{"key":"S0263574725000360_ref52","doi-asserted-by":"crossref","first-page":"3570","DOI":"10.1017\/S0263574724001553","article-title":"A semantic visual SLAM based on improved mask R-CNN in dynamic environment","volume":"42","author":"Zhang","year":"2024","journal-title":"Robotica"},{"key":"S0263574725000360_ref36","doi-asserted-by":"crossref","first-page":"573","DOI":"10.1109\/IROS.2012.6385773","volume-title":"2012 IEEE\/RSJ International Conference on Intelligent Robots and Systems","author":"Sturm","year":"2012"},{"key":"S0263574725000360_ref14","doi-asserted-by":"crossref","first-page":"670","DOI":"10.1007\/978-3-030-58452-8_39","volume-title":"Computer Vision\u2013ECCV 2020: 16th European Conference Part","author":"Dusmanu","year":"2020"},{"key":"S0263574725000360_ref28","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2017.2705103"},{"key":"S0263574725000360_ref23","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1017\/S026357471100049X","article-title":"Probabilistic map merging for multi-robot RBPF-SLAM with unknown initial poses","volume":"30","author":"Lee","year":"2012","journal-title":"Robotica"},{"key":"S0263574725000360_ref12","doi-asserted-by":"publisher","DOI":"10.1016\/j.asej.2022.102050"},{"key":"S0263574725000360_ref1","doi-asserted-by":"publisher","DOI":"10.1017\/S0263574722000625"},{"key":"S0263574725000360_ref27","doi-asserted-by":"crossref","first-page":"1147","DOI":"10.1109\/TRO.2015.2463671","article-title":"ORB-SLAM: A versatile and accurate monocular SLAM system","volume":"31","author":"Mur-Artal","year":"2015","journal-title":"IEEE Trans. Robot."},{"key":"S0263574725000360_ref20","first-page":"17408","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Johari","year":"2023"},{"key":"S0263574725000360_ref31","first-page":"3097","volume-title":"Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision","author":"Rosinol","year":"2023"},{"key":"S0263574725000360_ref46","first-page":"3","volume-title":"Robotics: Science and Systems","author":"Whelan","year":"2015"},{"key":"S0263574725000360_ref48","doi-asserted-by":"crossref","first-page":"145","DOI":"10.1145\/3653946.3653968","volume-title":"Proceedings of the 2024 7th International Conference on Machine Vision and Applications","author":"Xiao","year":"2024"},{"key":"S0263574725000360_ref49","first-page":"1281","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Yang","year":"2020"},{"key":"S0263574725000360_ref53","first-page":"10","article-title":"A BigBiGAN-based loop closure detection algorithm for indoor visual SLAM","volume":"2021","author":"Zhong","year":"2021","journal-title":"J. Electr. Comput. Eng."},{"key":"S0263574725000360_ref45","doi-asserted-by":"publisher","DOI":"10.1007\/s11517-022-02739-1"},{"key":"S0263574725000360_ref10","first-page":"5828","volume-title":"Scannet: Richly-annotated 3D reconstructions of indoor scenes","author":"Dai","year":"2017"},{"key":"S0263574725000360_ref26","doi-asserted-by":"crossref","first-page":"5218","DOI":"10.1109\/ICRA.2019.8794425","volume-title":"2019 International Conference on Robotics and Automation (ICRA)","author":"Loo","year":"2019"},{"key":"S0263574725000360_ref3","first-page":"24392","article-title":"Understanding and improving early stopping for learning with noisy labels","volume":"34","author":"Bai","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"S0263574725000360_ref15","doi-asserted-by":"crossref","first-page":"611","DOI":"10.1109\/TPAMI.2017.2658577","article-title":"Direct sparse odometry","volume":"40","author":"Engel","year":"2017","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"S0263574725000360_ref33","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2022.109115"},{"key":"S0263574725000360_ref41","first-page":"16558","article-title":"DROID-SLAM: Deep visual SLAM for monocular, stereo, and RGB-D cameras","volume":"34","author":"Teed","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"S0263574725000360_ref47","doi-asserted-by":"publisher","DOI":"10.1016\/j.robot.2019.03.012"},{"key":"S0263574725000360_ref34","doi-asserted-by":"crossref","first-page":"1452","DOI":"10.1017\/S0263574722001813","article-title":"Data augmentation by separating identity and emotion representations for emotional gait recognition","volume":"41","author":"Sheng","year":"2023","journal-title":"Robotica"},{"key":"S0263574725000360_ref50","unstructured":"[50] Yoshida, Y. and Miyato, T. , \u201cSpectral norm regularization for improving the generalizability of deep learning,\u201d (2017) arXiv preprint\u00a0arXiv:1705.10941"},{"key":"S0263574725000360_ref51","doi-asserted-by":"crossref","first-page":"2367","DOI":"10.1017\/S0263574724000754","article-title":"CPR-SLAM: RGB-D SLAM in dynamic environment using sub-point cloud correlations","volume":"42","author":"Yu","year":"2024","journal-title":"Robotica"},{"key":"S0263574725000360_ref55","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2020.2991614"},{"key":"S0263574725000360_ref19","doi-asserted-by":"publisher","DOI":"10.1016\/j.jksuci.2019.09.004"},{"key":"S0263574725000360_ref38","doi-asserted-by":"crossref","first-page":"1542","DOI":"10.1109\/SSCI.2018.8628742","volume-title":"2018 IEEE Symposium Series on Computational Intelligence (SSCI)","author":"Taylor","year":"2018"},{"key":"S0263574725000360_ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2014.6906584"},{"key":"S0263574725000360_ref7","doi-asserted-by":"crossref","first-page":"1874","DOI":"10.1109\/TRO.2021.3075644","article-title":"ORB-SLAM3: An accurate open-source library for visual, visual\u2013inertial, and multimap SLAM","volume":"37","author":"Campos","year":"2021","journal-title":"IEEE Trans. Robotics"},{"key":"S0263574725000360_ref25","doi-asserted-by":"crossref","first-page":"2169","DOI":"10.1017\/S0263574724000511","article-title":"A visual SLAM-based lightweight multi-modal semantic framework for an intelligent substation robot","volume":"42","author":"Li","year":"2024","journal-title":"Robotica"},{"key":"S0263574725000360_ref30","doi-asserted-by":"crossref","first-page":"1004","DOI":"10.1109\/TRO.2018.2853729","article-title":"VINS-Mono: A robust and versatile monocular visual-inertial state estimator","volume":"34","author":"Qin","year":"2018","journal-title":"IEEE Trans. Robot."},{"key":"S0263574725000360_ref42","doi-asserted-by":"publisher","DOI":"10.1017\/S0263574719000675"},{"key":"S0263574725000360_ref2","first-page":"8403262","article-title":"Object detection through modified YOLO neural network","volume":"2020","author":"Ahmad","year":"2020","journal-title":"Sci. Programming-NETH"},{"key":"S0263574725000360_ref18","doi-asserted-by":"crossref","first-page":"76","DOI":"10.46763\/JOE216.10076g","article-title":"Basic time series models in financial forecasting","volume":"6","author":"Gruevski","year":"2021","journal-title":"J. Econ."},{"key":"S0263574725000360_ref17","doi-asserted-by":"crossref","first-page":"105151","DOI":"10.1016\/j.engappai.2022.105151","article-title":"Ensemble deep learning: A review","volume":"115","author":"Ganaie","year":"2022","journal-title":"Eng. Appl. Artif. Intel."},{"key":"S0263574725000360_ref35","doi-asserted-by":"crossref","first-page":"746","DOI":"10.1007\/978-3-642-33715-4_54","volume-title":"Computer Vision\u2013ECCV 2012: 12th European Conference on Computer Vision","author":"Silberman","year":"2012"},{"key":"S0263574725000360_ref54","first-page":"1851","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Zhou","year":"2017"},{"key":"S0263574725000360_ref29","first-page":"1","volume-title":"Deep melanoma classification with k-fold cross-validation for process optimization","author":"Nie","year":"2020"},{"key":"S0263574725000360_ref9","doi-asserted-by":"crossref","first-page":"721","DOI":"10.1109\/LRA.2020.2965415","article-title":"DeepFactors: Real time probabilistic dense monocular SLAM","volume":"5","author":"Czarnowski","year":"2020","journal-title":"IEEE Robot. Autom. Lett."},{"key":"S0263574725000360_ref11","doi-asserted-by":"crossref","first-page":"1052","DOI":"10.1109\/TPAMI.2007.1049","article-title":"MonoSLAM: Real-time single camera SLAM","volume":"29","author":"Davison","year":"2007","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"S0263574725000360_ref22","first-page":"225","volume-title":"2007, 6th IEEE and ACM International Symposium on Mixed and Augmented Reality","author":"Klein","year":"2007"},{"key":"S0263574725000360_ref4","first-page":"7694","article-title":"Understanding batch normalization","volume":"31","author":"Bjorck","year":"2018","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"S0263574725000360_ref40","doi-asserted-by":"crossref","first-page":"402","DOI":"10.1007\/978-3-030-58536-5_24","volume-title":"Computer Vision\u2013ECCV 2020: 16th European Conference","author":"Teed","year":"2020"},{"key":"S0263574725000360_ref32","first-page":"3260","volume-title":"Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Sch\u00f6ps","year":"2017"},{"key":"S0263574725000360_ref44","doi-asserted-by":"publisher","DOI":"10.1109\/IROS45743.2020.9341801"},{"key":"S0263574725000360_ref43","first-page":"1761","volume-title":"Conference on Robot Learning","author":"Wang","year":"2021"},{"key":"S0263574725000360_ref6","first-page":"194","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Cai","year":"2021"},{"key":"S0263574725000360_ref8","doi-asserted-by":"crossref","first-page":"905","DOI":"10.1017\/S0263574713000131","article-title":"SLAM-based maneuverability strategy for unmanned car-like vehicles","volume":"31","author":"Cheein","year":"2013","journal-title":"Robotica"},{"key":"S0263574725000360_ref13","doi-asserted-by":"crossref","first-page":"11047","DOI":"10.1109\/IROS47612.2022.9981147","volume-title":"2022 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)","author":"Di Giammarino","year":"2022"}],"container-title":["Robotica"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.cambridge.org\/core\/services\/aop-cambridge-core\/content\/view\/S0263574725000360","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,23]],"date-time":"2025-07-23T02:51:50Z","timestamp":1753239110000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.cambridge.org\/core\/product\/identifier\/S0263574725000360\/type\/journal_article"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,31]]},"references-count":55,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2025,5]]}},"alternative-id":["S0263574725000360"],"URL":"https:\/\/doi.org\/10.1017\/s0263574725000360","relation":{},"ISSN":["0263-5747","1469-8668"],"issn-type":[{"type":"print","value":"0263-5747"},{"type":"electronic","value":"1469-8668"}],"subject":[],"published":{"date-parts":[[2025,3,31]]}}}