Visual localization based on images: methods, challenges, and trends (2019–2025)
Abstract
The article provides an overview of the domain of image-based localization systems, focusing on key methods, challenges, and current trends in development for the period 2019–2025. It considers the classification of approaches to visual localization, including traditional approaches based on local features and 3D structures, approaches based on image search (visual recognition of terrain), as well as the latest deep learning methods. It analyzes datasets, key industry challenges, and promising areas for further research.
References
Arandjelović, R., P. Gronat, A. Torii, T. Pajdla, and J. Sivic. 2016. "NetVLAD: CNN Architecture for Weakly Supervised Place Recognition." Proc. IEEE Conferenceon Computer Visionand Pattern Recognition (CVPR), 5297-307. https://arxiv.org/abs/1511.07247 https://doi.org/10.1109/CVPR.2016.572
Brachmann, E., A. Krull, S. Nowozin, etal. 2017. "DSAC - Differentiable RANSAC for Camera Localization." Proc. IEEE Conferenceon Computer Visionand Pattern Recognition (CVPR), 668-76. https://arxiv.org/abs/1611.05705 https://doi.org/10.1109/CVPR.2017.267
Brachmann, E., and C. Rother. 2019a. "Expert Sample Consensus Appliedto Camera Re-Localization." Proc. IEEE/CVF International Conferenceon Computer Vision (ICCV), 7525-34.https://arxiv.org/abs/1908.02484
Brachmann, E., and C. Rother. 2019b. "Neural-Guided RANSAC: Learning Whereto Sample Model Hypotheses." Proc. IEEE/CVF International Conferenceon Computer Vision (ICCV).https://arxiv.org/abs/1905.04132 https://doi.org/10.1109/ICCV.2019.00442
Chen, C., B. Wang, C. X. Lu, N. Trigoni, andA. Markham. 2024. "Deep Learning for Visual Localization and Mapping: A Survey." IEEE Transactionson Neural Networks and Learning Systems, aheadof print. https://arxiv.org/abs/2308.14039
Chen, Y. etal. 2022. Transformer Fusion: Point Cloud Guided Global Feature Fusionfor Image-Based Localization. https://arxiv.org/abs/2203.10722
DeTone, D., T. Malisiewicz, and A. Rabinovich. 2018. "SuperPoint: Self-Supervised Interest Point Detection and Description." Proc. IEEE/CVF Conferenceon Computer Visionand Pattern Recognition Workshops (CVPRW). https://arxiv.org/abs/1712.07629 https://doi.org/10.1109/CVPRW.2018.00060
Garg, S. etal. 2019. Semantic-geometric visual place recognition: a new perspective forrecon ciling opposing views geometric_visual_place_recognition_a_new_perspective_for_reconciling_opposing_views
https://doi.org/10.1177/0278364919839761
https://www.researchgate.net/publication/332280264_Semantic-
Kendall, A., M. Grimes, and R. Cipolla. 2015. "PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization." Proc. IEEE International Conferenceon Computer Vision (ICCV). https://arxiv.org/abs/1505.07427
https://doi.org/10.1109/ICCV.2015.336
Lin, Yen-Chenetal. 2021. iNeRF: Inverting Neural Radiance Fields for Pose Estimation. https://arxiv.org/abs/2012.05877 https://doi.org/10.1109/IROS51168.2021.9636708
Maddern, W., G. Pascoe, C. Linegar, and P. Newman. 2017. "1 Year, 1000 Km: The Oxford Robot Car Dataset." International Journal of Robotics Research (IJRR). https://ora.ox.ac.uk/objects/uuid%3Ac5266bad-e0f8-49f1-918e-1602ef935990
https://doi.org/10.1177/0278364916679498
Mapillary Research. 2021. Visual Place Recognition for Autonomous Driving. https://research.mapillary.com
Noh, H., A. Araujo, J. Sim, T. Weyand, and B. Han. 2017. "Large-Scale Image Retrieval with Attentive Deep Local Features." Proc. IEEE International Conferenceon Computer Vision (ICCV). https://arxiv.org/abs/1612.06321 https://doi.org/10.1109/ICCV.2017.374
Sarlin, P.-E., C. Cadena, R. Siegwart, and M. Dymczyk. 2019. "From Coarse to Fine: Robust Hierarchical Localizationat Large Scale." Proc. IEEE/CVF Conferenceon Computer Visionand Pattern Recognition (CVPR), 12716-25. https://arxiv.org/abs/1812.03506
Sarlin, P.-E., D. DeTone, T. Malisiewicz, and A. Rabinovich. 2020. "Super Glue: Learning Feature Matching with Graph Neural Networks." Proc. IEEE/CVF Conferenceon Computer Vision and Pattern Recognition (CVPR), 4937-46. https://arxiv.org/abs/1911.11763 https://doi.org/10.1109/CVPR42600.2020.00499
Copyright (c) 2025 Іван Ільницький, Петро Пукач (Автор)
This work is licensed under a Creative Commons Attribution 4.0 International License.
