Visual Place Recognition: A Survey
Open accessThe DOI matches its Crossref record; 233 of the 239 references it cites were found in a registry or an index.
Abstract
Visual place recognition is a challenging problem due to the vast range of ways in which the appearance of real-world places can vary. In recent years, improvements in visual sensing capabilities, an ever-increasing focus on long-term mobile robot autonomy, and the ability to draw on state-of-the-art research in other disciplines-particularly recognition in computer vision and animal navigation in neuroscience-have all contributed to significant advances in visual place recognition systems. This paper presents a survey of the visual place recognition research landscape. We start by introducing the concepts behind place recognition-the role of place recognition in the animal kingdom, how a “place” is defined in a robotics context, and the major components of a place recognition system. Long-term robot operations have revealed that changing appearance can be a significant factor in visual place recognition failure; therefore, we discuss how place recognition solutions can implicitly or explicitly account for appearance change within the environment. Finally, we close with a discussion on the future of visual place recognition, in particular with respect to the rapid advances being made in the related fields of deep learning, semantic scene understanding, and video description.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Items 201–233 of 233
McManus, C., Furgale, P., & Barfoot, T. D. (2013) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2013.04.008 · Cited by 34 · cited 1×
“[202] used scanning laser-rangefinders to create “camera-like” images that”
MethodIncremental topo-metric SLAM using vision and robot odometry
Bazeille, S., & Filliat, D. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5979908 · Cited by 32
Autonomous Multisensor Calibration and Closed‐loop Fusion for SLAM
Jacobson, A., Chen, Z., & Milford, M. J. (2014) · Journal of Field Robotics · doi:10.1002/rob.21500 · Cited by 31 · cited 1×
“If multiple streams of data are available, a voting scheme [3], [5], [79], [96], [175] can be used.”
MethodSelf-citationEfficient on-line data summarization using extremum summaries
Girdhar, Y., & Dudek, G. (2012) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2012.6224657 · Cited by 31
Environment selection and hierarchical place recognition
Mohan, M., Gálvez‐López, D., Monteleoni, C., & Sibley, G. (2015) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2015.7139966 · Cited by 30 · cited 1×
“[143] selected the most likely environment using cooccurent feature matrices.”
MethodIncremental unsupervised topological place discovery
Murphy, L., & Sibley, G. (2014) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2014.6907022 · Cited by 30 · cited 1×
“[55] combined dynamic vocabulary building [56] and incremental topic modeling [57] to continually learn new topological places in an environment, and Volkov et al.”
BackgroundShatkay, H., & Kaelbling, L. P. (2002) · Journal of Artificial Intelligence Research · doi:10.1613/jair.874 · Cited by 29
An Online Sparsity-Cognizant Loop-Closure Algorithm for Visual Navigation
Latif, Y., Huang, G. P., Leonard, J. J., & Neira, J. (2014) · Robotics: Science and Systems · doi:10.15607/rss.2014.x.036 · Cited by 29 · cited 2×
“[19] used topological information to formulate place recognition as a sparse convex L1-minimization problem and applied efficient homotopy methods [146] to provide loop closure hypotheses.”
MethodSelf-citationRobust outdoor visual localization using a three‐dimensional‐edge map
Nuske, S. T., Roberts, J. M., & Wyeth, G. F. (2009) · Journal of Field Robotics · doi:10.1002/rob.20306 · Cited by 28
BRIEF-Gist - Closing the loop by simple means
Sünderhauf, N., & Protzel, P. (2011) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2011.6048590 · Cited by 27
Self-citationVanishing points and three-dimensional lines from omni-directional video
Bosse, M. C., Rikoski, R. J., Leonard, J. J., & Teller, S. J. (2003) · The Visual Computer · doi:10.1007/s00371-003-0205-3 · Cited by 26
Self-citationMultiple map hypotheses for planning and navigating in non-stationary environments
Morris, T., Dayoub, F., Corke, P., Wyeth, G. F., & Upcroft, B. (2014) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2014.6907255 · Cited by 26 · cited 4×
“The systems presented above [220], [226], [227], [232] were designed for metric systems.”
BackgroundInfluentialSelf-citationCoresets for visual summarization with applications to loop closure
Volkov, M., Rosman, G., Feldman, D., Fisher, J. S., & Rus, D. L. (2015) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2015.7139704 · Cited by 25 · cited 1×
“[58] used coresets”
Visual precis generation using coresets
Paul, R., Feldman, D., Rus, D. L., & Newman, P. M. (2014) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2014.6907021 · Cited by 24 · cited 1×
“However, Paul and Newman [60], [62], [185] presented an iterative learning mechanism to generate a representative set of the true distribution of the appearance of the world.”
BackgroundSelf-citationUsing Gateways To Build A Route Map
Kortenkamp, D., Baker, L., & Weymouth, T. E. (2005) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.1992.602138 · Cited by 22 · cited 1×
“where the appearance changes significantly, called a “gateway” [47].”
BackgroundRobot environment modeling via principal component regression
Vlassis, N., & Kröse, B. (2003) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.1999.812758 · Cited by 22 · cited 1×
“Early examples of appearance-based probabilistic localization used Gaussians to represent probability [177], or a mixture of Gaussians combined with expectation maximization [178], or a Gaussian kernel [179] with Parzen smoothing [104].”
MethodHow was your day? Online visual workspace summaries using incremental clustering in topic space
Paul, R., Rus, D. L., & Newman, P. M. (2012) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2012.6224762 · Cited by 20 · cited 5×
“For example, Ho and Newman [84] used Harris affine regions [85], Murillo et al. [86] and Cummins and Newman [87] Fig.”
BackgroundInfluentialSelf-citationImage Sequence Partitioning for outdoor mapping
Korrapati, H., Courbon, J., Mezouar, Y., & Martinet, P. (2012) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2012.6224892 · Cited by 17 · cited 1×
“[53] used image sequencing partitioning techniques to group visually similar images together as topological”
MethodCapping computation time and storage requirements for appearance-based localization with CAT-SLAM
Maddern, W., Milford, M. J., & Wyeth, G. F. (2012) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2012.6224622 · Cited by 15 · cited 1×
“Computation time is thus proportional to the number of particles, not the size of the environment [145].”
BackgroundSelf-citationLong exposure localization in darkness using consumer cameras
Milford, M. J., Turner, I., & Corke, P. (2013) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2013.6631105 · Cited by 15 · cited 1×
“Whole-image descriptors have been used in systems such as SeqSLAM and others [118], [119], [197] that demonstrate robustness against environmental change.”
MethodSelf-citationDong, J. F., Wijesoma, S., & Shacklock, A. P. (2007) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2007.4399077 · Cited by 15 · cited 1×
“Elements of a scene that are moving when the robot observes them must be detected and may also be removed [229], [230].”
BackgroundVisual localization in highly crowded urban environments
Hafez, A. H. A., Singh, M., Krishna, K. M., & Jawahar, C. V. (2013) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2013.6696749 · Cited by 14
Robust Online Change-point Detection in Video Sequences
Tsechpenakis, G., Metaxas, D. N., Neidle, C., & Hadjiliadis, O. (2006) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2006.176 · Cited by 14
Appearance-based segmentation of indoors/outdoors sequences of spherical views
Chapoulie, A., Rives, P., & Filliat, D. (2013) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2013.6696614 · Cited by 13
Unsupervised online learning of condition-invariant images for place recognition
Lowry, S. M., Wyeth, G. F., & Milford, M. J. (2014) · QUT ePrints (Queensland University of Technology) · Cited by 13
Self-citationA novel method for analysing lighting variance
Ross, P., English, A., Ball, D., Upcroft, B., Wyeth, G. F., & Corke, P. (2013) · QUT ePrints (Queensland University of Technology) · Cited by 11
Self-citationProbabilistic Appearance-Based Mapping and Localization Using Visual Features
Garcia‐Fidalgo, E., & Ortiz, A. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-38628-2_33 · Cited by 11 · cited 1×
“Garcia-Fidalgo and Rodriguez [182] used the observation likelihood that relates the number of feature matches between two images to the overall number of features in the image, scaled by a normalizing constant.”
MethodMixture conditional density estimation with the EM algorithm
Vlassis, N. (1999) · doi:10.1049/cp:19991213 · Cited by 8 · cited 1×
“Early examples of appearance-based probabilistic localization used Gaussians to represent probability [177], or a mixture of Gaussians combined with expectation maximization [178], or a Gaussian kernel [179] with Parzen smoothing [104].”
MethodA method to quantify a descriptor's illumination variance
Ross, P., English, A., Ball, D., & Corke, P. (2014) · QUT ePrints (Queensland University of Technology) · Cited by 7
Self-citationSelf-help: Seeking out perplexing images for ever improving topological mapping
Paul, R., & Newman, P. M. (2013) · The International Journal of Robotics Research · doi:10.1177/0278364913509859 · Cited by 5 · cited 1×
“However, Paul and Newman [60], [62], [185] presented an iterative learning mechanism to generate a representative set of the true distribution of the appearance of the world.”
BackgroundSelf-citationSpecial Issue on Long-Term Autonomy
Barfoot, T., Kelly, J. W., & Sibley, G. (2013) · The International Journal of Robotics Research · doi:10.1177/0278364913511182 · Cited by 5
Exploratory analysis of time-lapse imagery with fast subset PCA
Abrams, A., Feder, E., & Pless, R. B. (2011) · Proceedings · doi:10.1109/wacv.2011.5711523 · Cited by 4
Shatkay, H., & Kaelbling, L. P. (2002) · Cited by 4
Research works printed but not linked: 6
- neira, 0, Mobile robot localization and map building using monocular vision, Proc 5th Symp Intell Robot Syst, 275
- lynch, 1960, The Image of the City
- 10.1177/0278364914551008 · DOI
- 10.1177/027836402761412467 · doi:10.1177/027836402761412467 · Paper page
- 10.21236/ADA576491 · doi:10.21236/ada576491 · Paper page
- sim, 2005, Vision-based SLAM using the Rao-Blackwellised particle filter, Proc IJCAI Workshop Reason Uncertainty Robot
Cited by
Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age
Cadena, C., Carlone, L., Carrillo, H., Latif, Y., Scaramuzza, D., Neira, J., Reid, I., & Leonard, J. J. (2016) · IEEE Transactions on Robotics · doi:10.1109/tro.2016.2624754 · Cited by 3,596
“[160], and some of its challenges are discussed in Section III.”
Self-citationPast, present, and future of simultaneous localization and mapping: Toward the robust-perception age
Cadena, C., Carlone, L., Carrillo, H., Latif, Y., Scaramuzza, D., Neira, J., Reid, I. D., & Leonard, J. J. (2016) · Zurich Open Repository and Archive (University of Zurich) · doi:10.5167/uzh-138892 · Cited by 3,048
Self-citationSimultaneous Localization and Mapping: A Survey of Current Trends in Autonomous Driving
Bresson, G., Alsayed, Z., Yu, L., & Glaser, S. (2017) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2017.2749181 · Cited by 885
“focus on how to deal with seasonal or weather changes and we refer our readers to this survey [93] for a detailed view of this field, independently of loop closure.”
Autonomous vehicle perception: The technology of today and tomorrow
Van Brummelen, J., O’Brien, M., Gruyer, D., & Najjaran, H. (2018) · Transportation Research Part C: Emerging Technologies · doi:10.1016/j.trc.2018.02.012 · Cited by 869
Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions
Sattler, T., Maddern, W., Toft, C., Torii, A., Hammarstrand, L., Stenborg, E., Safari, D., Okutomi, M., Pollefeys, M., Sivic, J., et al. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00897 · Cited by 699
“They are often used for place recognition [1, 17, 41, 55, 66, 69] and loop-closure detection [20,25,48].”
Background
Metadata
- Authors
- Lowry, S., Sünderhauf, N., Newman, P., Leonard, J. J., Cox, D., Corke, P., & Milford, M. J.
- Affiliations
- Queensland University of Technology (AU)
- University of Oxford (GB)
- MIT Computer Science and Artificial Intelligence Laboratory (US)
- Harvard University (US)
- ORCID
- Venue
- IEEE Transactions on Robotics
- Published
- 2016-02
- Volume
- 32
- Issue
- 1
- Pages
- 1-19
- ISSN
- 1552-3098 (ISSN-L)
- 1941-0468
- OpenAlex
- W2284029970
- Cited by
- 1,079
- References
- 239
- Citations by year
- 2026: 47 · 2025: 63 · 2024: 111 · 2023: 112 · 2022: 109 · 2021: 148 · 2020: 156 · 2019: 132 · 2018: 100 · 2017: 72 · 2016: 29
- FWCI
- 97.28
- Type
- journal article
- Language
- en
- Open access
- green
- purl.org
- Licence
- Locations
- IEEE Transactions on Robotics (publishedVersion)
- QUT ePrints (Queensland University of Technology) (submittedVersion)
- Funders
- ARC Future Fellowship (FT140101229)
- Australian Centre for Robotic Vision
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Spatial Cognition and Navigation
- Keywords
- visual place recognition, place recognition, appearance changes, animal navigation, semantic scene understanding, audio description, deep learning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.