FAB-MAP: Probabilistic Localization and Mapping in the Space of Appearance
The DOI matches its Crossref record; 54 of the 57 references it cites were found in a registry or an index.
Abstract
This paper describes a probabilistic approach to the problem of recognizing places based on their appearance. The system we present is not limited to localization, but can determine that a new observation comes from a previously unseen place, and so augment its map. Effectively this is a SLAM system in the space of appearance. Our probabilistic approach allows us to explicitly account for perceptual aliasing in the environment—identical but indistinctive observations receive a low probability of having come from the same place. We achieve this by learning a generative model of place appearance. By partitioning the learning problem into two parts, new place models can be learned online from only a single observation of a place. The algorithm complexity is linear in the number of places in the map, and is particularly suitable for online loop closure detection in mobile robotics.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Adhi Harmoko, S., M.Komp, Jacquard, J. M., Zuse, K., & Eniac (2005) · Cited by 18,035
Proceedings of IEEE Conference on Computer Vision and Pattern Recognition
(1993) · doi:10.1109/cvpr.1993 · Cited by 17,667
Rayward-Smith, V. J. (1991) · Journal of the Operational Research Society · doi:10.1057/jors.1991.155 · Cited by 16,821
Object recognition from local scale-invariant features
Lowe, D. G. (1999) · Proceedings of the Seventh IEEE International Conference on Computer Vision · doi:10.1109/iccv.1999.790410 · Cited by 16,472 · cited 2×
“Local feature schemes consist of a region-of-interest detector combined with a descriptor of the local region, SIFT (Lowe 1999) being a popular example.”
BackgroundSURF: Speeded Up Robust Features
Bay, H., Tuytelaars, T., & Van Gool, L. (2006) · Lecture Notes in Computer Science · doi:10.1007/11744023_32 · Cited by 14,955
Advances in neural information processing systems 7
(1996) · Computers & Mathematics with Applications · doi:10.1016/s0898-1221(96)90463-0 · Cited by 14,281
(1999) · Cited by 12,854
Grossman, P. (2009) · Discrete Mathematics for Computing · doi:10.1007/978-0-230-37405-8_1 · Cited by 8,834
Video Google: a text retrieval approach to object matching in videos
Sivic, & Zisserman (2003) · Proceedings Ninth IEEE International Conference on Computer Vision · doi:10.1109/iccv.2003.1238663 · Cited by 6,367 · cited 4×
“Finally, we map regions of descriptor space to visual words as suggested by Sivic and Zisserman (2003).”
MethodInfluentialGumbel, E. J. (1958) · Columbia University Press eBooks · doi:10.7312/gumb92958 · Cited by 5,567
Research works printed but not linked: 27
- Bay, H., 2006, Proceedings of the 9th European Conference on Computer Vision, Graz, Austria, 13, 404
- Bowling, M., Twenty-First National Conference on Artificial Intelligence
- Bowling, M., Proceedings of the International Symposium of Robotics Research
- Dementiev, R., Proceedings 3rd International Conference on Theoretical Computer Science
- Ferreira, F., Proceedings IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems
- Gutmann, J., Proceedings of the IEEE International Symposium on Computational Intelligence in Robotics and Automation (CIRA)
- Ho, K., Proceedings of the European Conference on Mobile Robotics
- Ho, K., Proceedings of the International Conference on Computational Intelligence
- Johnson, A., 1997, Spin-Images: a representation for 3-D surface matching
- Li, F., IEEE International Conference on Robotics and Automation
- Lowe, D.G., Proceedings of the 7th International Conference on Computer Vision
- Meila, M., ICML '99: Proceedings of the Sixteenth International Conference on Machine Learning
- Meila2-Predoviciu, M., 1999, Learning with mixtures of trees
- Mozos, O., Proceedings of the IEEE International Conference on Robotics and Automation
- Newman, P., Proceedings of the IEEE International Conference on Robotics and Automation
- Newman, P.M., Proceedings of the IEEE International Conference on Robotics and Automation (ICRA)
- Nister, D., 2006, Proceedings of the Conference on Computer Vision and Pattern Recognition, 2, 2161
- Ramos, F.T., Proceedings of the IJCAI Workshop on Reasoning with Uncertainty in Robotics
- Ranganathan, A., Proceedings of International Conference on Robotics and Automation
- Schindler, G., Proceedings IEEE Conference on Computer Vision and Pattern Recognition
- Silpa-Anan, C., Proceedings of the Australian Conference on Robotics and Automation
- Silpa-Anan, C., Proceedings of the Workshop on Omnidirectional Vision
- Sim, R., Proceedings of the IEEE International Conference on Intelligent Robots and Systems
- Sivic, J., Proceedings of the International Conference on Computer Vision
- Srebro, N., Proceedings of the 17th Annual Conference on Uncertainty in Artificial Intelligence (UAI-01)
- Torralba, A., ICCV '03: Proceedings of the Ninth IEEE International Conference on Computer Vision
- Wang, J., Proceedings of the IEEE International Conference on Robotics and Automation
Cited by
Items 601–700 of 1,002
Adaptive iterative closest keypoint
Ekekrantz, J., Pronobis, A., Folkesson, J. B., & Jensfelt, P. (2013) · European Conference on Mobile Robots · doi:10.1109/ecmr.2013.6698824 · Cited by 13
“This has laid the foundation for FAB-MAP [30] and its follow-ups.”
BackgroundRelative navigation approach for vision-based aerial GPS-denied navigation
Leishman, R. C., McLain, T. W., & Beard, R. W. (2013) · International Conference on Unmanned Aircraft Systems (ICUAS) · doi:10.1109/icuas.2013.6564707 · Cited by 13
Irie, K., Yoshida, T., & Tomono, M. (2010) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2010.5654401 · Cited by 13
“Recently, appearance-based localization methods which are robust to changes in lighting have been proposed [14][15].”
BackgroundUnsupervised 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
Vision-based navigation method for a local maneuvering of the autonomous underwater vehicle
Бобков, В. А., Mel’man, S. V., Kudrashov, A., & Scherbatyuk, A. (2017) · IEEE Underwater Technology (UT) · doi:10.1109/ut.2017.7890304 · Cited by 15
“Later it was further developed in a number of papers, for example, in the often cited paper [6] where the system FAB-MAP with an improved probabilistic model based on the bag-of-wordsrepresentations is described.”
BackgroundEvaluation Criteria for Inside-Out Indoor Positioning Systems Based on Machine Learning
Löffler, C., Riechel, S., Fischer, J., & Mutschler, C. (2018) · International Conference on Indoor Positioning and Indoor Navigation · doi:10.1109/ipin.2018.8533862 · Cited by 18
A Novel Loop Closure Detection Approach Using Simplified Structure for Low-Cost LiDAR
Ye, Q., Shi, P., Xu, K., Gui, P., & Zhang, S. (2020) · Sensors · doi:10.3390/s20082299 · Cited by 13
A Comparison of Similarity Measures for Localization with Passive RFID Fingerprints
Vorst, P., & Zell, A. (2010) · Cited by 13
Hidden view synthesis using real-time visual SLAM for simplifying video surveillance analysis
Christopher Mei, A., Sommerlade, E., Sibley, G., Newman, P. M., & Reid, I. D. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5980093 · Cited by 14
“approaches scale well with the size of the environment and in this work we use the FABMAP algorithm [4] available online.”
MethodSelf-citationHou, Y., Zhang, H., Zhou, S., & Zou, H. (2017) · Mobile Information Systems · doi:10.1155/2017/8104386 · Cited by 13
“A popular baseline algorithm among traditional approaches is FAB-MAP [2].”
MethodA High Dynamic Range vision approach to outdoor localization
Kiyoshi, I. R. I. E., Yoshida, T., & Tomono, M. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5980281 · Cited by 12
Automatic image scaling for place recognition in changing environments
Pepperell, E., Corke, P., & Milford, M. J. (2015) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2015.7139316 · Cited by 12
“Precision refers to the proportion of returned frame pairs that were correct (within 40 m of ground truth – similar to past studies [5], [24]; though the majority were within 20 m), and recall is the proportion of total correct frame pairs that were returned.”
ResultVisual Odometry and Place Recognition Fusion for Vehicle Position Tracking in Urban Environments
Ouerghi, S., Boutteau, R., Savatier, X., & Tlili, F. (2018) · Sensors · doi:10.3390/s18040939 · Cited by 12
“BoW representation has been first used by the popular algorithm FAB-MAP [8] that relies on extracting scale-invariant image keypoints and descriptors.”
MethodGeometry-Constrained Scale Estimation for Monocular Visual Odometry
Zhang, H., Wang, X., Yin, X., Du, M., Liu, C., & Chen, Q. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3093771 · Cited by 13
QuickLogS: A Quick Log Parsing Algorithm based on Template Similarity
Fang, L., Di, X., Liu, X., Qin, Y., Ren, W., & Ding, Q. (2021) · IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom) · doi:10.1109/trustcom53373.2021.00148 · Cited by 14
“We compare the templates from each group, and the template vectors are obtained through the bag of words [34].”
MethodTransformer-based descriptors with fine-grained region supervisions for visual place recognition
Wang, Y., Qiu, Y., Cheng, P., & Zhang, J. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.110993 · Cited by 12
Towards Fast Image-Based Localization on a City-Scale
Sattler, T., Leibe, B., & Kobbelt, L. P. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-34091-8_9 · Cited by 12
“Current state-of-the-art methods such as [6, 11, 12] try to adapt the SLAM approach to increasingly large scenes for real-time localization.”
BackgroundMethodGeometry matters: Place recognition in 2D range scans using Geometrical Surface Relations
Himstedt, M., & Maehle, E. (2015) · European Conference on Mobile Robots · doi:10.1109/ecmr.2015.7324185 · Cited by 12
“The latter estimates a relative transformation of the query scan with respect to the matching one, whereas appearance based approaches solely match nodes of topological graphs [1].”
BackgroundMethodBio-inspired visual memory for robot cognitive map building and scene recognition
Rebai, K., Azouaoui, O., & Achour, N. (2012) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2012.6385493 · Cited by 12
“This idea is largely used in image retrieval methods [22], object and place recognition in robotics [13].”
MethodInfluentialEvent-driven loop closure in multi-robot mapping
Vidal‐Calleja, T., Berger, C., & Lacroix, S. (2009) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2009.5354335 · Cited by 12
“Such matches must be searched between submaps that are likely to match, which can either be selected on the basis of their position estimate 3, or by applying loop closure detection approaches such as the one based on image indexing [13] – the latter approaches having never been applied to multirobot systems to our kn…”
BackgroundCombining Odometry and Visual Loop-Closure Detection for Consistent Topo-Metrical Mapping
Bazeille, S., & Filliat, D. (2010) · RAIRO - Operations Research · doi:10.1051/ro/2010021 · Cited by 12
“The vision sensor is suitable for many robotic applications such as user interaction or object and place recognition [1][5], and has also been used in many Simultaneous Localization And Mapping (SLAM) solutions (e.g. [6][7][8]).”
MethodMobile robot localization using multiple observations based on place recognition and GPS
Saito, T., & KURODA, Y. (2013) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2013.6630776 · Cited by 12
“FAB-MAP uses only image information as appearance based place recognition method [6], and it can be formulated as a recursive Bayes estimation : p(Li|Zk) = p(Zk|Li)p(Li|Zk−1) p(Zk|Zk−1) (1) Here, Li is an location, Zk is an observation of a local scene appearance captured at time k. Zk is all…”
BackgroundMethodAutomatic calibration of camera sensor networks based on 3D texture map information
Ji, Y., Yamashita, A., & Asama, H. (2016) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2016.09.015 · Cited by 13
LexToMap: lexical-based topological mapping
Rangel, J. C., Martínez-Gómez, J., García-Varea, I., & Cazorla, M. (2016) · Advanced Robotics · doi:10.1080/01691864.2016.1261045 · Cited by 13
“[22] More concretely,[21] provides a way to detect loop clousure, but the proposed system needs to learn the visual features in the environment.”
BackgroundMethodBiologically-inspired visual place recognition with adaptive multiple scales
Fan, C., Chen, Z., Jacobson, A., Hu, X., & Milford, M. J. (2017) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2017.07.015 · Cited by 13
“The place recognition performance is compared with other state ‐ of ‐ the ‐ art place recognition algorithms, including multi ‐ scale (MS) place recognition [3], SeqSLAM [18] and FABMAP [9, 10].”
MethodResultEfficient Indoor Positioning with Visual Experiences via Lifelong Learning
Wen, H., Clark, R., Wang, S., Lu, C. X., Du, B., Hu, W. S., & Trigoni, N. (2018) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2018.2852645 · Cited by 12
“This means although L-Learning only processes a tiny portion of information comparing to FAB-MAP and NaviGlass, it is able to achieve nearly the same accuracy.”
MethodInfluentialDeterministic Optimality for Robust Vehicle Localization Using Visual Measurements
Jiao, Y., Wang, Y., Ding, X., Wang, M., & Xiong, R. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3053574 · Cited by 15
Loop Closure Detection based on Image Covariance Matrix Matching for Visual SLAM
Ying, T., Yan, H., Li, Z., Shi, K., & Feng, X. (2021) · International Journal of Control Automation and Systems · doi:10.1007/s12555-020-0730-0 · Cited by 12
A Hybrid Loop Closure Detection Method Based on Brain-Inspired Models
Li, J., Tang, H., & Yan, R. (2022) · IEEE Transactions on Cognitive and Developmental Systems · doi:10.1109/tcds.2022.3152910 · Cited by 12
“Thus, image-to-image matching methods are commonly used for LCD [30], [31].”
MethodPairwise Probabilistic Voting: Fast Place Recognition without RANSAC
Johns, E., & Yang, G. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-10605-2_33 · Cited by 12
“Whilst such constraints offer powerful verification of global geometric consistency, they are expensive to compute due to the need for generating multiple transformation hypotheses in the RANSAC algorithm, and large-scale real-time recognition tasks are often forced to forego 3D geometry entirely [2][7].”
BackgroundOutdoor visual localization with a hand-drawn line drawing map using FastSLAM with PSO-based mapping
Matsuo, K., & Miura, J. (2012) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2012.6385506 · Cited by 12
“Previous works on visual localization in outdoor environments can roughly be divided into two categories: landmark-based ones [18], [7], [5], [8] and view-based ones [2], [14].”
BackgroundAdaptive, real-time visual simultaneous localization and mapping
Clipp, B., Zach, C., Lim, J., Frahm, J., & Pollefeys, M. (2009) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2009.5403054 · Cited by 12
“Another possible approach to detecting visually similar areas was developed by Cummins and Murray in [4].”
MethodAn integrated model of autonomous topological spatial cognition
Karaoğuz, H., & Bozma, H. I. (2015) · Autonomous Robots · doi:10.1007/s10514-015-9514-4 · Cited by 12
“1 Note that this definition differs appearance-based or topological SLAMmethods where each location is considered separately as a place (Cummins and Newman 2008; Newman et al. 2009a; Konolige et al. 2010) or a representative location (key-place) is selected after grouping visual data from different locations (Murphy a…”
BackgroundBeyond a Shadow of a Doubt: Place Recognition with Colour-Constant Images
MacTavish, K., Paton, M., & Barfoot, T. D. (2016) · Springer tracts in advanced robotics · doi:10.1007/978-3-319-27702-8_13 · Cited by 12
“The FAB-MAP algorithm and its extensions [5, 4, 6] have been extensively tested and widely used; in particular, FAB-MAP 2.”
MethodInfluentialPose Graph Compression for Laser-Based SLAM
Stachniss, C., & Kretzschmar, H. (2016) · Springer tracts in advanced robotics · doi:10.1007/978-3-319-29363-9_16 · Cited by 13
A Self-Calibrating, Vision-Based Navigation Assistant
Koch, O., & Teller, S. J. (2008) · Cited by 12
Accurate Vision-based Vehicle Localization using Satellite ImageryPreprint
Chu, H., Mei, H., Bansal, M., & Walter, M. R. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1510.09171 · Cited by 13
“The FAB-MAP algorithm by Cummins and Newman [3, 5] is designed to address challenges that arise as a result of perceptual aliasing.”
MethodVisual Localization Using Global Visual Features and Vanishing Points.
Saurer, O., Fraundorfer, F., & Pollefeys, M. (2010) · Cited by 11
Past and State-of-the-Art SLAM Technologies
Song, J., & Hwang, S.-Y. (2014) · Journal of Institute of Control Robotics and Systems · doi:10.5302/j.icros.2014.14.9024 · Cited by 11
“비전분야에서는 최근 장소를 인식하는 방법으로 FAB-MAP (Fast Appearance-Based Mapping) [50]이 주로 사용되 고 있다.”
Visual localization and loop closing using decision trees and binary features
Schlegel, D., & Grisetti, G. (2016) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2016.7759679 · Cited by 11
“According to the reported results, FabMap pushes the BoW approaches to its limits and allows to quickly retrieve similar images on very large datasets [11].”
ResultAn End-To-End Siamese Convolutional Neural Network for Loop Closure Detection in Visual Slam System
Liu, H., Zhao, C., Huáng, W., & Shi, W. (2018) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2018.8462426 · Cited by 11
“In the experiments, we evaluate the proposed end-to-end siamese convolutional neural network on New College dataset [6], City Centre dataset [6] and TUM open dataset [22].”
MethodVision and Wi-Fi fusion in probabilistic appearance-based localization
Nowakowski, M., Joly, C., Dalibard, S., Garcia, N., & Moutarde, F. (2020) · The International Journal of Robotics Research · doi:10.1177/0278364920910485 · Cited by 12
“In our work, these parameters are set to the values specified in Cummins and Newman (2008).”
BackgroundMethodInfluentialPark, C.-S., Chae, H.-W., & Song, J.-B. (2020) · International Journal of Control Automation and Systems · doi:10.1007/s12555-019-0891-x · Cited by 11
LIV-LAM: LiDAR and Visual Localization and Mapping
Radmanesh, R., Wang, Z., Chipade, V. S., Tsechpenakis, G., & Panagou, D. (2020) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc45564.2020.9148037 · Cited by 12
“The front-end can detect these loop closures based on visual recognition (sensory vision), or by LiDAR as in [20], [3], [11].”
MethodБобков, В. А., Kudryashov, A. P., & Inzartsev, A. (2022) · Remote Sensing · doi:10.3390/rs14205123 · Cited by 11
Location Learning for AVs: LiDAR and Image Landmarks Fusion Localization with Graph Neural Networks
Kang, Q., She, R., Wang, S., Tay, W. P., Navarro, D. N., & Hartmannsgruber, A. (2022) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc55140.2022.9922179 · Cited by 12
“In the literature of landmark map construction, [20] uses Edge Boxes [21] to detect a bounding box (bbox) around a patch that contains a larger number of internal contours compared to the number of contours exiting from the box, which indicates the presence of an intelligible object in the enclosed…”
MethodLu, F., Zhang, L., Dong, S., Chen, B., & Yuan, C. (2023) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48891.2023.10160734 · Cited by 11
“Some aggregation algorithms, such as Bag of Words (BoW) [13] and Vector of Locally Aggregated Descriptors (VLAD) [4], [5], were used to aggregate local features like SURF [14], [15].”
MethodRobust Imaging Sonar-based Place Recognition and Localization in Underwater Environments
Kim, H., Kang, G., Jeong, S., Ma, S., & Cho, Y. (2023) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48891.2023.10161518 · Cited by 11
“For aerial and ground robotics, there are several well-known methods with vision [1, 2] and light detection and ranging (LiDAR)-based [3, 4, 5] sensors.”
MethodA framework for robust cognitive spatial mapping
Pronobis, A., Sjöö, K., Aydemir, A., Bishop, A. N., & Jensfelt, P. (2009) · Cited by 11
A hull census transform for scene change detection and recognition towards topological map building
Wang, M.-L., & Lin, H. (2010) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2010.5652817 · Cited by 11
“[4] proposed a bag-of-words based loop-close detection.”
Vision-based Markov localization across large perceptual changes
Naseer, T., Suger, B., Ruhnke, M., & Burgard, W. (2015) · European Conference on Mobile Robots · doi:10.1109/ecmr.2015.7324181 · Cited by 11
“In the last experiment, we chose the New College dataset [6], a typical single run mapping scenario with identical database and query sets.”
MethodSaito, M., Kiuchi, K., Shimizu, S., Yokota, T., Fujino, Y., Saito, T., & KURODA, Y. (2014) · Journal of Robotics and Mechatronics · doi:10.20965/jrm.2014.p0185 · Cited by 11
Sparse optimization for robust and efficient loop closing
Latif, Y., Huang, G. P., Leonard, J. J., & Neira, J. (2017) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2017.03.016 · Cited by 11
“In particular, the FAB-MAP [13] is a probabilistic appearance-based approach using visual BoW for place recognition, and was shown to work robustly over trajectories up to 1000 km. Similarly, the BinaryBoW (BBoW)-based method [21] detects the FAST keypoints [38] and employs a variation of the BRIEF descriptors [6] to…”
BackgroundMethodResultInfluentialVisual Global Localization with a Hybrid WNN-CNN Approach
Forechi, A., Oliveira-Santos, T., Badue, C. S., & Souza, A. F. D. (2018) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2018.8489335 · Cited by 12
Beyond SIFT using binary features in Loop Closure Detection
Han, L., Zhou, G., Xu, L., & Fang, L. (2017) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2017.8206261 · Cited by 11
“Image [14] 199 and 300 are captured at different places, but shared a high similarity score.”
BackgroundMethodInfluentialGao, P., & Zhang, H. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i06.6604 · Cited by 11
“A category of these methods focused on using visual appearance information to represent and match places, e.g., based on local features (Cummins and Newman 2008; Mur-Artal and Tard´os 2014) and visual holistic scenes (Naseer et al. 2014; Latif et al. 2014).”
MethodFang, Y., Wang, K., Cheng, R., & Yang, K. (2020) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv47402.2020.9304612 · Cited by 11
“Sometimes holistic images can be presented by local features [23] [24] while sometimes by global descriptors [25] [26].”
BackgroundDiscriminative and semantic feature selection for place recognition towards dynamic environments
Tian, Y., Miao, J., Wu, X., Yue, H., Liu, Z., & Chen, W. (2021) · Pattern Recognition Letters · doi:10.1016/j.patrec.2021.11.014 · Cited by 11
Loop-Closure Detection in Urban Scenes for Autonomous Robot Navigation
Maffra, F., Teixeira, L., Chen, Z., & Chli, M. (2017) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2017.00048 · Cited by 11
“Full feature-based comparisons can be computationally expensive, and therefore most of the underlying structure and geometry between features is generally ignored, such as in [5].”
BackgroundMethodShi, Y., Ji, S.-P., Shao, X.-W., Yang, P., Wu, W.-B., Shi, Z.-C., & Shibasaki, R. (2015) · Image and Vision Computing · doi:10.1016/j.imavis.2015.06.002 · Cited by 11
“Using camera motion sequences to calculate the probability of loop closure through the location prior, the CAT-SLAM [3] algorithm determines more correct loop closures than the first generation of FAB-MAP.”
MethodInfluentialProbabilistic 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
“Although many works assume the availability of omnidirectional images, and typically operate offline [1,2,3,4], many others make use of cheaper/easier to acquire ordinary imaging configurations [5,6,7,8].”
BackgroundMethodA review of path planning and mapping technologies for autonomous mobile robot systems
Dhiman, N. K., Deodhare, D., & Khemani, D. (2012) · Bangalore Annual Compute Conference · doi:10.1145/2459118.2459121 · Cited by 11
“Highly Scalable Appearance-Only SLAM FAB-MAP 2.0.”
MethodInfluentialIla, V., Porta, J. M., & Andrade‐Cetto, J. (2011) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2011.02.010 · Cited by 14
“Keywords: State recovery, Kalman filter, Information filter, Pose SLAM, Hierarchical SLAM”
MethodGirdhar, Y., & Dudek, G. (2010) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/robot.2010.5509464 · Cited by 13
“The seminal work by Cummins and Newman [7] on loop closure involves ability to identify the common geographic origin of images, which might not be completely similar; thereby producing a meaningful clustering of images, which could be used to generate summaries.”
BackgroundAdaptive appearance based loop-closing in heterogeneous environments
Majdik, A. L., Galvez-Lopez, D., Lazea, G., & Castellanos, J. A. (2011) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2011.6094537 · Cited by 12
“In [2] the results of the visual similarity algorithm are improved by learning the co-appearance probability of the visual words in the scenes.”
BackgroundMethodInfluentialA moving observer in a three-dimensional world
Glennerster, A. (2016) · Philosophical Transactions of the Royal Society B Biological Sciences · doi:10.1098/rstb.2015.0265 · Cited by 13
Utilizing WiFi signals for improving SLAM and person localization
Kudo, T., & Miura, J. (2017) · IEEE/SICE International Symposium on System Integration (SII) · doi:10.1109/sii.2017.8279260 · Cited by 11
“Image-based loop closure detection [5], [6], [7] is also popular.”
BackgroundPlaneLoc: Probabilistic global localization in 3-D using local planar features
Wietrzykowski, J., & Skrzypczyński, P. (2019) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2019.01.008 · Cited by 11
Flow-based SLAM: From geometry computation to learning
Yan, Z., & Zha, H. (2019) · Virtual Reality & Intelligent Hardware · doi:10.1016/j.vrih.2019.09.001 · Cited by 12
Zou, Z., Zheng, C., Yuan, C., Zhou, S., Xue, K., & Zhang, F. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3483032 · Cited by 11
“It’s noteworthy that handcrafted visual local descriptors necessitate collaboration with other tools [2], [3] for loop closure retrieval.”
BackgroundMethodEDTformer: An Efficient Decoder Transformer for Visual Place Recognition
Jin, T., Lu, F. H., Hu, S., Yuan, C., & Liu, Y. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3559084 · Cited by 23
“In early VPR research, global features were developed by aggregating the hand-crafted local features (e.g., SIFT [40] and SURF [41], [42]), employing some classical aggregation algorithms, such as Bag of Words [43], Fisher Vector [44] and Vector of Locally Aggregated Descriptors (VLAD) [45].”
AutoInspect: Toward Long-Term Autonomous Inspection and Monitoring
Staniaszek, M., Flatscher, T., Rowell, J., Niu, H., Liu, W., You, Y., Gadd, M., Mattamala, M., Schutz, A., Martini, D. D., et al. (2025) · IEEE transactions on field robotics. · doi:10.1109/tfr.2025.3586831 · Cited by 17
“Our system uses FAST-BRIEF feature point detection and description [77], [78], visual odometry [79], and place recognition with FAB-MAP [80].”
A New Approach to Train Convolutional Neural Networks for Real-Time 6-DOF Camera Relocalization
Esfahani, M. A., Wu, K., Yuan, S., & Wang, H. (2018) · International Conference on Control and Automation · doi:10.1109/icca.2018.8444299 · Cited by 13
“relocalization [16], [17] also achieved satisfactory results in relocalizing camera position in limited and discretized places.”
BackgroundRSS-Net: Weakly-Supervised Multi-Class Semantic Segmentation with FMCW Radar
Kaul, P., Martini, D. D., Gadd, M., & Newman, P. M. (2020) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv47402.2020.9304674 · Cited by 81
“We use the ground truth pose dataset described in the recently released Oxford Radar RobotCar Dataset [24] which is computed by an optimisation using Global Positioning System (GPS), robust Visual Odometry (VO) [25], and visual loop closures from FAB-MAP [26].”
MethodSelf-citationA Bayesian Approach for Determining Player Abilities in Football
Whitaker, G. A., Silva, R., Edwards, D., & Kosmidis, I. (2020) · Journal of the Royal Statistical Society Series C (Applied Statistics) · doi:10.1111/rssc.12454 · Cited by 10
Schubert, S., Neubert, P., & Protzel, P. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3052446 · Cited by 10
“For example, given the images from the database (or another representative training set), FabMap [13] learns statistics about feature occurrences in order to determine the most descriptive features.”
BackgroundMethodInfluentialKong, D., Li, X., Hu, Y., Xu, Q., Wang, A., & Hu, W. (2022) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2022.3169849 · Cited by 10
“FAB-MAP [18] is the first study to combine handcrafted features and BoW encoding for place recognition.”
BackgroundLoop Closure Detection via Locality Preserving Matching With Global Consensus
Ma, J., Zhang, K., & Jiang, J. (2022) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2022.105926 · Cited by 11
“Often, off-line methods train visual dictionaries in advance, according to which BoW vectors are calculated directly to represent images [4], [22]–[25].”
BackgroundMethodLocalization based on multiple visual-metric maps
Sujiwo, A., Takeuchi, E., Morales, L. Y., Akai, N., Ninomiya, Y., & Edahiro, M. (2017) · IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI) · doi:10.1109/mfi.2017.8170431 · Cited by 10
“Early efforts for vision-based mapping and localization were centered in topological methods, such as the FABMAP [8].”
MethodZeng, F., Jacobson, A., Smith, D., Boswell, N., Peynot, T., & Milford, M. J. (2019) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2019.2932579 · Cited by 10
Zhao, C., Ding, R., & Key, H. L. (2019) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2019.8802931 · Cited by 11
“great success in static scenes [3, 4, 5].”
TCL: Tightly Coupled Learning Strategy for Weakly Supervised Hierarchical Place Recognition
Shen, Y., Wang, R., Zuo, W., & Zheng, N. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3141663 · Cited by 10
“…the recommended evaluation of datasets, ground truth matches are obtained by given binary matrix from authors for New College [18] and City Centre [19], 25m translation error for Pitts30k [10], [12], 25m translation and 40 ◦ orientation error for MSLS [14], and 5 closet corresponding frames (i.e.…”
MethodInfluentialDanish Airs and Grounds: A Dataset for Aerial-to-Street-Level Place Recognition and Localization
Vallone, A., Warburg, F., Hansen, H. N., Hauberg, S., & Civera, J. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3187491 · Cited by 14
Image-based localization using prior map database and Monte Carlo Localization
Kim, H., Oh, T., Lee, D., & Myung, H. (2014) · International Conference on Ubiquitous Robots and Ambient Intelligence · doi:10.1109/urai.2014.7057440 · Cited by 10
“Keywords – Image-based Localization, Image Feature, Prior Map Database, MCL”
BackgroundTopological map induction using neighbourhood information of places
Werner, F., Sitte, J., & Maire, F. (2012) · Autonomous Robots · doi:10.1007/s10514-012-9276-1 · Cited by 10
“…successful approach for loop closure detection using visual data association is FAB-MAP which employs a probabilistic approach for image matching based on a bag-of-words model for performing localisation on trajectories up to 100 km in length (Cummins and Newman 2008, 2009; Posner et al. 2009).”
MethodUsing feature scale change for robot localization along a route
Vardy, A. (2010) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2010.5649557 · Cited by 10
Beyond RatSLAM: Improvements to a biologically inspired SLAM system
Sünderhauf, N., & Protzel, P. (2010) · IEEE International Conference on Emerging Technologies and Factory Automation (ETFA) · doi:10.1109/etfa.2010.5641280 · Cited by 10
“It therefore bears some resemblance to the recently published FAB-MAP system [3] [4] but uses a much more simple descriptor to recognize familiar scenes.”
MethodConstructing Topological Maps using Markov Random Fields and Loop-Closure Detection
Anati, R., & Daniilidis, K. (2009) · Cited by 10
Survey Registration for Long‐Term Natural Environment Monitoring
Griffith, S., & Pradalier, C. (2016) · Journal of Field Robotics · doi:10.1002/rob.21664 · Cited by 12
“A dictionary of representative features are used in some approaches (e.g., FabMAP [Cummins and Newman, 2008]).”
MethodBruce, J., Jacobson, A., & Milford, M. J. (2017) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2017.2724146 · Cited by 10
“Single-frame approaches for localization in a sensory stream attempt to find a match in a database for each incoming frame, and this has been demonstrated using image description techniques such as global “gist” representations [11], feature vocabularies ([12], [13]), and neural networks [14].”
MethodA Combined RGB and Depth Descriptor for SLAM with Humanoids
Sheikh, R., Obwald, S., & Bennewitz, M. (2018) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8593768 · Cited by 10
“The bag-of-words descriptor of the new keyframe is added to the test set of FAB-MAP for future loop closure matches only if 10 frames have passed since the last added keyframe.”
MethodInfluentialMaltar, J., Marković, I., & Petrovic, I. K. (2020) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2020.103598 · Cited by 12
“In the context of visual place recognition SURF [4] is used in FAB-MAP [12], while ORB features [6, 42] are used in ORB-SLAM [34].”
MethodProbabilistic Appearance-Invariant Topometric Localization With New Place Awareness
Xu, M., Fischer, T., Sünderhauf, N., & Milford, M. J. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3096745 · Cited by 19
Trajectory bundle estimation For perception-driven planning
Bachrach, A. (2013) · DSpace@MIT (Massachusetts Institute of Technology) · Cited by 16
Applying information theory to efficient SLAM
Chli, M. (2009) · Spiral (Imperial College London) · doi:10.25560/5634 · Cited by 18
Visual place recognition using landmark distribution descriptorsPreprint
Panphattarasap, P., & Calway, A. D. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1608.04274 · Cited by 10
“To aid robustness, these techniques are often incorporated within some form of temporal integration, the probabilistic FAB-MAP method [5, 6] being the most well known.”
MethodTextGeo-SLAM: A LiDAR SLAM With Text Semantics and Geometric-Constraint-Based Loop Closure
Chen, S., Li, C., Jiang, Q., Zhuang, X., Zhang, B., Zhou, B., & Li, Q. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3517699 · Cited by 9
“Among the image-based methods, the most renowned is the bag of words (BOW) method [27], [28] which represents images as different word combinations from the dictionary.”
Self-Organizing Incremental Associative Memory-Based Robot Navigation
Tangruamsub, S., Kawewong, A., Tsuboyama, M., & Hasegawa, O. (2012) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.e95.d.2415 · Cited by 9
“In addition, to concentrate on SLAM, most previous works have subsumed guided exploration; the robot is controlled by a human to collect data from an environment [1]–[3].”
BackgroundPlace Recognition Based Visual Localization Using LBP Feature and SVM
Qiao, Y., Cappelle, C., & Ruichek, Y. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-27101-9_30 · Cited by 10
A LoCATe‐based visual place recognition system for mobile robotics and GPGPUs
Bampis, L., Chatzichristofis, S. A., Iakovidou, C., Amanatiadis, A. A., Boutalis, Y. S., & Gasteratos, A. C. (2017) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.4146 · Cited by 9
“Each bin represents a color as follows: (0) Black, (1) Grey, (2) White, (3) Dark Red, (4) Red, (5) Light Red, (6) Dark Orange, (7) Orange, (8) Light Orange, (9) Dark Yellow, (10) Yellow, (11) Light Yellow, (12) Dark Green, (13) Green, (14) Light Green, (15) Dark Cyan, (16) Cyan, (17) Light Cyan, (18) Dark Blue, (19) B…”
Metadata
- Authors
- Cummins, M., & Newman, P.
- Affiliations
- University of Oxford (GB)
- Corresponding author
- Cummins, Mark
- Newman, Paul
- ORCID
- Newman, Paul
- Venue
- The International Journal of Robotics Research
- Publisher
- SAGE Publications
- Published
- 2008-06
- Volume
- 27
- Issue
- 6
- Pages
- 647-665
- ISSN
- 0278-3649 (ISSN-L)
- 1741-3176
- OpenAlex
- W2144824356
- Cited by
- 1,472
- References
- 57
- Citations by year
- 2026: 28 · 2025: 43 · 2024: 60 · 2023: 59 · 2022: 119 · 2021: 122 · 2020: 103 · 2019: 117 · 2018: 105 · 2017: 127 · 2016: 90 · 2015: 103 · 2014: 72 · 2013: 83 · 2012: 71
- FWCI
- 116.69
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- journals.sagepub.com/page/policies/text-and-data-mining-license (tdm)
- Locations
- The International Journal of Robotics Research (publishedVersion)
- Funders
- Engineering and Physical Sciences Research Council
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Keywords
- place recognition, perceptual aliasing, loop closure detection, mobile robots
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.