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 101–200 of 1,002
Navigating, Recognizing and Describing Urban Spaces With Vision and Lasers
Newman, P., Sibley, G., Smith, M., Cummins, M., Harrison, A., Mei, C., Posner, I., Shade, R., Schroeter, D., Murphy, L., et al. (2009) · The International Journal of Robotics Research · doi:10.1177/0278364909341483 · Cited by 134
“Our loop closure detection component, “FAB-MAP” (Fast Appearance Based Mapping) [13], [15], [14] is probabilistic and solely appearance-based.”
MethodSelf-citationA Comprehensive Review on Autonomous Navigation
Nahavandi, S., Alizadehsani, R., Nahavandi, D., Mohamed, S. M. K., Mohajer, N., Rokonuzzaman, M., & Hossain, I. (2025) · ACM Computing Surveys · doi:10.1145/3727642 · Cited by 190
“Researchers have developed loop closure mechanisms [74, 106, 271] based on bag-of-words (BoWs) approach to remedy the high computational load of descriptors such as SIFT and SURF.”
SVIn2: An Underwater SLAM System using Sonar, Visual, Inertial, and Depth Sensor
Rahman, S., Li, A. Q., & Rekleitis, I. M. (2019) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros40897.2019.8967703 · Cited by 170
“FAB-MAP [33], [34] is an appearance-based method to recognize places in a probabilistic framework.”
MethodAll-environment visual place recognition with SMART
Pepperell, E., Corke, P., & Milford, M. J. (2014) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2014.6907067 · Cited by 131
DisLocation: Scalable Descriptor Distinctiveness for Location Recognition
Arandjelović, R., & Zisserman, A. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16817-3_13 · Cited by 140
“Such a system is useful for geotagging personal photos [1], mobile augmented reality [2], robot localization [3], or to aid automatic 3D reconstruction [4].”
BackgroundFAB-MAP 3D: Topological mapping with spatial and visual appearance
Paul, R., & Newman, P. (2010) · 2010 IEEE International Conference on Robotics and Automation · doi:10.1109/robot.2010.5509587 · Cited by 125
“The next step is building the vocabulary model by constructing the Chow-Liu tree by a procedure outlined in [9].”
BackgroundMethodInfluentialSelf-citationWhere Is Your Place, Visual Place Recognition?
Garg, S., Fischer, T., & Milford, M. J. (2021) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2021/603 · Cited by 174
“Notable exceptions include system papers where VPR is one of many components, and a specific downstream task is considered, such as [24].”
BackgroundMethodKeypoint design and evaluation for place recognition in 2D lidar maps
Bosse, M. C., & Zlot, R. (2009) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2009.07.009 · Cited by 132
“If instead, we had an efficient global appearance-based test, then we could quickly solve t he problem independent of the integrated map uncertainty [4].”
MethodLategahn, H., & Stiller, C. (2014) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2014.2298492 · Cited by 132
“Beyond that, it is related to place recognition [3], [4], [8], [20], [27], [28] sometimes referred to as loop-closure detection in the SLAM literature.”
BackgroundMethodMap-Based Probabilistic Visual Self-Localization
Brubaker, M. A., Geiger, A., & Urtasun, R. (2015) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2015.2453975 · Cited by 137
BoW3D: Bag of Words for Real-Time Loop Closing in 3D LiDAR SLAM
Cui, Y., Chen, X., Zhang, Y., Dong, J., Wu, Q., & Zhu, F. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3221336 · Cited by 140
“In recent years, a variety of image-based methods [1], [4], [5], [15], [16], [17] have been proposed.”
BackgroundMethodFactoring the Mapping Problem: Mobile Robot Map-building in the Hybrid Spatial Semantic Hierarchy
Beeson, P., Modayil, J., & Kuipers, B. (2010) · The International Journal of Robotics Research · doi:10.1177/0278364909100586 · Cited by 119
“For example, Cummins and Newman [2008] discuss closing loops over kilometers of travel, where small rotational errors lead to large positional errors, and the correct loop closure may never be considered by odometry-based solutions.”
BackgroundMethodInfluentialVisual SLAM for underwater vehicles: A survey
Zhang, S., Zhao, S., An, D. U., Liu, J., Wang, H., Yu, F., Li, D., & Zhao, R. (2022) · Computer Science Review · doi:10.1016/j.cosrev.2022.100510 · Cited by 136
Online discovery and maintenance of time series motifs
Mueen, A., & Keogh, E. (2010) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/1835804.1835941 · Cited by 116
“In an effort to verify this utility, we use the "New College" dataset [7], where a set of 2,146 images have been collected by a moving robot.”
MethodCAT-SLAM: probabilistic localisation and mapping using a continuous appearance-based trajectory
Maddern, W., Milford, M., & Wyeth, G. (2012) · The International Journal of Robotics Research · doi:10.1177/0278364912438273 · Cited by 118
“These are determined using the mean field approximation (Jordan et al., 1999), or the random sampling method used by Cummins and Newman (2008) and presented in Equation (10).”
BackgroundMethodInfluentialExploiting Sparse Semantic HD Maps for Self-Driving Vehicle Localization
Ma, W.-C., Tartavull, I., Bârsan, I. A., Wang, S., Bai, M., Máttyus, G., Homayounfar, N., Lakshmikanth, S. K., Pokrovsky, A., & Urtasun, R. (2019) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros40897.2019.8968122 · Cited by 138
“With this problem in mind, [9, 21, 27] proposed an image-based localization technique that is to some degree invariant to appearance changes.”
BackgroundRole of Deep Learning in Loop Closure Detection for Visual and Lidar SLAM: A Survey
Arshad, S., & Kim, G.-W. (2021) · Sensors · doi:10.3390/s21041243 · Cited by 115
“In [8], a probabilistic framework is presented, FABMAP, for appearance-based place recognition and loop closing.”
MethodResultNeuroSLAM: a brain-inspired SLAM system for 3D environments
Yu, F., Shang, J., Hu, Y., & Milford, M. J. (2019) · Biological Cybernetics · doi:10.1007/s00422-019-00806-9 · Cited by 131
Semantics for Robotic Mapping, Perception and Interaction: A Survey
Garg, S., Sünderhauf, N., Dayoub, F., Morrison, D., Cosgun, A., Carneiro, G., Wu, Q., Chin, T.-J., Reid, I., Gould, S., et al. (2020) · Foundations and Trends® in Robotics · doi:10.1561/2300000059 · Cited by 105
SeqNet: Learning Descriptors for Sequence-Based Hierarchical Place Recognition
Garg, S., & Milford, M. J. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3067633 · Cited by 137
“FAB-MAP [1], robust local [11], and global feature representations [9], [12], efficient retrieval [13]–[15], and biologicallyinspired techniques [16].”
BackgroundInformation-theoretic compression of pose graphs for laser-based SLAM
Kretzschmar, H., & Stachniss, C. (2012) · The International Journal of Robotics Research · doi:10.1177/0278364912455072 · Cited by 109
“Since then, a large variety of approaches to minimize the error in pose graphs have been proposed, for example, Estrada et al. (2005), Folkesson et al. (2005), Grisetti et al. (2010), Konolige and Agrawal (2008), and Ni et al. (2007). Most of these approaches to SLAM do not provide a means to effectively prune the gra…”
BackgroundInfluentialEfficient Large‐scale Three‐dimensional Mobile Mapping for Underground Mines
Zlot, R., & Bosse, M. C. (2014) · Journal of Field Robotics · doi:10.1002/rob.21504 · Cited by 117
“A fundamental difference between our keypoint voting algorithm and other place recognition approaches such as bag-of-words (Cummins and Newman, 2008; GálvezLópez and Tardós, 2012; Steder, Ruhnke, Grzonka, and Burgard, 2011) and NDT (Magnusson et al.”
MethodA Constant-Time Efficient Stereo SLAM System
Mei, C., Sibley, G., Cummins, M., Newman, P., & Reid, I. (2009) · Procedings of the British Machine Vision Conference 2009 · doi:10.5244/c.23.54 · Cited by 107
Self-citationSummary Maps for Lifelong Visual Localization
Mühlfellner, P., Bürki, M., Bosse, M., Derendarz, W., Philippsen, R., & Furgale, P. (2016) · Journal of Field Robotics · doi:10.1002/rob.21595 · Cited by 110
“A more general scheme could be based on appearance-based global localization [see, e.g., Cummins & Newman (2008, 2011)] or a GNSS estimate.”
MethodConvolutional Neural Network-Based Robot Navigation Using Uncalibrated Spherical Images
Ran, L., Zhang, Y., Zhang, Q., & Yang, T. (2017) · Sensors · doi:10.3390/s17061341 · Cited by 118
Automatic Visual Bag-of-Words for Online Robot Navigation and Mapping
Nicosevici, T., & Garcia, R. (2012) · IEEE Transactions on Robotics · doi:10.1109/tro.2012.2192013 · Cited by 106
DiSCO: Differentiable Scan Context With Orientation
Xu, X., Yin, H., Chen, Z., Li, Y., Wang, Y., & Xiong, R. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3060741 · Cited by 114
“Unseen place rejection stated in [18], [30] is also important to prevent false data associations.”
BackgroundLocNet: Global Localization in 3D Point Clouds for Mobile Vehicles
Yin, H., Tang, L., Ding, X., Wang, Y., & Xiong, R. (2018) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2018.8500682 · Cited by 119
“Most image based localization methods employed bag of words for description of a frame [3].”
Method3D LiDAR-Based Global Localization Using Siamese Neural Network
Yin, H., Wang, Y., Ding, X., Tang, L., Huang, S., & Xiong, R. (2019) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2019.2905046 · Cited by 113
“Most visual localization methods employed BoW to achieve descriptions of images [2], [17].”
BackgroundMethodVisual loop closure detection with a compact image descriptor
Liu, Y., & Zhang, H. (2012) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2012.6386145 · Cited by 104
“The FAB-MAP developed by Cummins and Newman [5], [6] is considered an important step in appearance-based SLAM.”
BackgroundMethodInfluentialOutdoor place recognition in urban environments using straight lines
Lee, J. H., Lee, S., Zhang, G., Lim, J., Chung, W. K., & Suh, I. H. (2014) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2014.6907675 · Cited by 100
Beyond Cross-view Image Retrieval: Highly Accurate Vehicle Localization Using Satellite Image
Shi, Y., & Li, H. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.01650 · Cited by 176
“It is solved by ground-toground (G2G) image matching [3, 8, 9, 13, 14, 20, 29, 34, 36, 40, 43, 58].”
BackgroundTowards life-long visual localization using an efficient matching of binary sequences from images
Arroyo, R., Alcantarilla, P. F., Bergasa, L. M., & Romera, E. (2015) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2015.7140088 · Cited by 102
“According to this, FAB-MAP [1] can be considered as the milestone in visual topological localization methods for detecting loop closures.”
MethodFusion and binarization of CNN features for robust topological localization across seasons
Arroyo, R., Alcantarilla, P. F., Bergasa, L. M., & Romera, E. (2016) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2016.7759685 · Cited by 102
“FAB-MAP is tested using the OpenFABMAP toolbox [33].”
BackgroundMethodInfluentialAutomatic Relocalization and Loop Closing for Real-Time Monocular SLAM
Williams, B., Klein, G., & Reid, I. D. (2011) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2011.41 · Cited by 101
Locus: LiDAR-based Place Recognition using Spatiotemporal Higher-Order Pooling
Vidanapathirana, K., Moghadam, P., Harwood, B., Zhao, M., Sridharan, S., & Fookes, C. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9560915 · Cited by 116
“Bag-of-Words (BoW) [16] and its higher-order variants (Vector of Locally Aggregated Descriptors (VLAD) [17], NetVLAD [10], Fisher Vector (FV) [18]) are popular meth-”
BackgroundAitken, J. M., Evans, M. H., Worley, R., Edwards, S., Zhang, R., Dodd, T. J., Mihaylova, L. S., & Anderson, S. (2021) · IEEE Access · doi:10.1109/access.2021.3115981 · Cited by 116
“The fast appearance-based mapping (FAB-MAP) algorithm [36] developed a visual appearance recognition algorithm using a bag-of-words (BOW) approach modified”
MethodSingle-View Place Recognition under Seasonal ChangesPreprint
Olid, D., Fácil, J. M., & Civera, J. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.06516 · Cited by 122
“Two of the most relevant among these techniques are FAB-MAP [10] and DBoW [4].”
MethodLeveraging Deep Visual Descriptors for Hierarchical Efficient Localization
Sarlin, P.-E., Debraine, F., Dymczyk, M., Siegwart, R., & Cadena, C. (2018) · Repository for Publications and Research Data (ETH Zurich) · doi:10.3929/ethz-b-000318818 · Cited by 128
Multi-camera visual SLAM for autonomous navigation of micro aerial vehicles
Yang, S., Scherer, S., Yi, X., & Zell, A. (2017) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2017.03.018 · Cited by 101
“Although we have seen successful applications of visual SLAM on ground vehicles (Cummins and Newman, 2008; Strasdat et al., 2011), there are more challenges in using visual SLAM to enable autonomous navigation of MAVs.”
MethodHierarchical Place Recognition for Topological Mapping
Garcia‐Fidalgo, E., & Ortiz, A. (2017) · IEEE Transactions on Robotics · doi:10.1109/tro.2017.2704598 · Cited by 93
“[6] M. Cummins and P. Newman, “Appearance-only SLAM at large scale with FAB-MAP 2.0,” Int.”
MethodInfluentialFast-SeqSLAM: A fast appearance based place recognition algorithm
Siam, S. M., & Zhang, H. (2017) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2017.7989671 · Cited by 98
“Bag-of-Words, a popular technique initially developed for efficient text retrieval and later extended to image matching [8], has been extensively used for place recognition [9], [10], [11].”
BackgroundMagnusson, M., Andreasson, H., Nüchter, A., & Lilienthal, A. J. (2009) · Journal of Field Robotics · doi:10.1002/rob.20314 · Cited by 89
“Cummins and Newman (2007, 2008a, 2008b, 2009) have published several articles on visual loop detection using their FAB-MAP method.”
BackgroundMethodInfluentialLocalization in Urban Environments Using a Panoramic Gist Descriptor
Murillo, A. C., Singh, G., Košecká, J., & Guerrero, J. J. (2012) · IEEE Transactions on Robotics · doi:10.1109/tro.2012.2220211 · Cited by 96
“For example, Cummins and Newman presented an appearance-based localization method, handling extremely large trajectories with their FAB-MAP approach [11], and Valgren and Lilienthal [12] evaluated localization across different season variations.”
InfluentialPrivacy Preserving Image-Based Localization
Speciale, P., Schönberger, J. L., Kang, S. B., Sinha, S. N., & Pollefeys, M. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.00564 · Cited by 117
“Localizing a device within a scene by computing the camera pose from an image is a fundamental problem in computer vision, with high relevance in applications such as robotics [16, 19, 65], augmented/mixed reality (AR/MR) [36, 46], and structure from motion (SfM) [25, 61, 63].”
BackgroundIndoor scene recognition through object detection
Espinace, P., Kollar, T., Soto, Á., & Roy, N. (2010) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/robot.2010.5509682 · Cited by 93
“In terms of robotics, besides the fact that some of the already mentioned methods are applied to this field, extensive work has been done in the case of topological localization using visual landmarks [22] [23].”
MethodGuo, X., Hu, J., Chen, J., Deng, F., & Lam, T. L. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3058935 · Cited by 100
CricaVPR: Cross-Image Correlation-Aware Representation Learning for Visual Place Recognition
Lu, F. H., Lan, X., Zhang, L., Jiang, D., Wang, Y., & Yuan, C. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.01587 · Cited by 160
“…Recognition: The early VPR approaches typically represent place images by global features that are computed using aggregation algorithms, such as Bag of Words [3] and Vector of Locally Aggregated Descriptors (VLAD) [4, 30, 35, 39, 59], to aggregate the hand-crafted descriptors like SURF [6, 16].”
MethodRobust Place Recognition With Stereo Sequences
Cadena, C., Galvez-López, D., Tardos, J. D., & Neira, J. (2012) · IEEE Transactions on Robotics · doi:10.1109/tro.2012.2189497 · Cited by 85
“Our place recognition system can be summarized in algorithm 1.”
BackgroundAccelerated appearance-only SLAM
Cummins, M., & Newman, P. M. (2008) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/robot.2008.4543473 · Cited by 95
“We have noted related ideas in other fields [9], however we believe our approach is novel in this context.”
BackgroundSelf-citationIBuILD: Incremental bag of Binary words for appearance based loop closure detection
Khan, S., & Wollherr, D. (2015) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2015.7139959 · Cited by 91
“Dataset Description Camera position Image size # Images Malaga6L [21] Outdoor, slightly dynamic Frontal 1024 x 768 3474 City Centre [22] Outdoor, urban, dynamic Lateral 640 x480 1237 Lip6 Indoor [7] Indoor, static Frontal 240x192 388 Lip6 Outdoor [7] Outdoor, slightly dynamic Frontal 240x192 531”
BackgroundLearning Semantic Maps from Natural Language Descriptions
Walter, M., Hemachandra, S., Homberg, B., Tellex, S., & Teller, S. (2013) · Robotics: Science and Systems IX · doi:10.15607/rss.2013.ix.004 · Cited by 84
“A number of solutions exist that identify loop closures based upon visual appearance [19, 20] and local metric structure [21], among others.”
BackgroundHybrid CNN-Transformer Features for Visual Place Recognition
Wang, Y., Qiu, Y., Cheng, P., & Zhang, J. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3212434 · Cited by 98
“Early methods mainly employ handcrafted features [21], [22], [23], [24] combined with Bag of Words (BoW) [25], Fisher Vector (FV) [26] and VLAD [10].”
MethodKeyframe-based real-time camera tracking
Dong, Z., Zhang, G., Jia, J., & Bao, H. (2009) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2009.5459273 · Cited by 87
“There are approaches to employ the invariant features for object and location recognition [27, 24, 25, 7, 8, 1].”
MethodWebb, B. (2019) · Journal of Experimental Biology · doi:10.1242/jeb.188094 · Cited by 83
“This typically combines dead reckoning with landmark recognition in a joint probabilistic estimate of robot and landmark positions (Cummins and Newman, 2008; Davison et al., 2007; Engel et al., 2014; Mur-Artal and Tardos, 2017).”
BackgroundMulti-Process Fusion: Visual Place Recognition Using Multiple Image Processing Methods
Hausler, S., Jacobson, A., & Milford, M. J. (2019) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2019.2898427 · Cited by 96
“FAB-MAP [7] is a noteworthy example of an appearance-only localization approach, which uses a recursive Bayesian model to estimate location over a series of past visual feature observations.”
MethodLearning to Localize Using a LiDAR Intensity MapPreprint
Bârsan, I. A., Wang, S., Pokrovsky, A., & Urtasun, R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.10902 · Cited by 105
“However, these approaches either require computationally demanding online 3D map rendering [15] or lack robustness to visual apperance changes due to the time of day, weather, and seasons [3, 4].”
BackgroundGeometric Loss Functions for Camera Pose Regression with Deep Learning
Kendall, A., & Cipolla, R. (2017) · arXiv (Cornell University) · doi:10.17863/cam.26496 · Cited by 73
Feature Co-occurrence Maps: Appearance-based localisation throughout the day
Johns, E., & Yang, G.-Z. (2013) · 2013 IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2013.6631024 · Cited by 85
MinkLoc++: Lidar and Monocular Image Fusion for Place Recognition
Komorowski, J., Wysoczańska, M., & Trzcinski, T. P. (2021) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn52387.2021.9533373 · Cited by 120
Superpixel-based appearance change prediction for long-term navigation across seasons
Neubert, P., Sünderhauf, N., & Protzel, P. (2014) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2014.08.005 · Cited by 93
“In a first experiment we evaluated the performance of FAB-MAP [6] (using the openFAB-MAP implementation) on the Nordland dataset.”
BackgroundMethodInfluentialSatellite image-based localization via learned embeddings
Kim, D.-K., & Walter, M. R. (2017) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2017.7989239 · Cited by 95
Fast loop-closure detection using visual-word-vectors from image sequences
Bampis, L., Amanatiadis, A. A., & Gasteratos, A. C. (2017) · The International Journal of Robotics Research · doi:10.1177/0278364917740639 · Cited by 85
“In their work, sequence-based descriptors were assessed for a variety of different but fixed-sized image groups, while the matching was achieved using a FAB-MAP-based probabilistic scheme.”
MethodInfluentialSLAM Overview: From Single Sensor to Heterogeneous Fusion
Chen, W., Zhou, C., Shang, G., Wang, X., Li, Z., Xu, C., & Hu, K. (2022) · Remote Sensing · doi:10.3390/rs14236033 · Cited by 91
Memory management for real-time appearance-based loop closure detection
Labbé, M., & Michaud, F. (2011) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2011.6094602 · Cited by 100
“66 GHz Intel Core i7, we conducted tests with the following community data sets: NewCollege and CityCentre data sets [3], and Lip6Indoor and”
BackgroundMethodInfluentialTowards dense moving object segmentation based robust dense RGB-D SLAM in dynamic scenarios
Wang, Y., & Huang, S. (2014) · International Conference on Control, Automation, Robotics and Vision (ICARCV) · doi:10.1109/icarcv.2014.7064596 · Cited by 81
“It is based on this method that we propose a practical way to do dense moving object segmentation for robust dense SLAM in dynamic scenarios.”
MethodGOSMatch: Graph-of-Semantics Matching for Detecting Loop Closures in 3D LiDAR data
Zhu, Y., Ma, Y., Chen, L. F., Liu, C., Ye, M., & Li, L. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341299 · Cited by 82
Indoor localization and visualization using a human-operated backpack system
Liu, T., Carlberg, M. A., Chen, G., Chen, J. H. K., Kua, J., & Zakhor, A. (2010) · International Conference on Indoor Positioning and Indoor Navigation · doi:10.1109/ipin.2010.5646820 · Cited by 93
“Our algorithm is a modified version of Newman’s FABMAP [12], a probabilistic approach of recognizing places via appearance.”
BackgroundMethodInfluentialMahajan, H. B., Uke, N., Pise, P. D., Shahade, M., Dixit, V. G., Bhavsar, S., & Deshpande, S. D. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-14253-5 · Cited by 87
“Sensors can be noisy and behaviors are non-deterministic and so uncertain [16].”
BackgroundMethodVisual topological SLAM and global localization
Angeli, A., Doncieux, S., Meyer, J.-A., & Filliat, D. (2009) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/robot.2009.5152501 · Cited by 85
“…size, low weight and low energy requirements of a simple camera make it an integrated sensor that can be easily embedded on most mobile robots, while vision provides a rich qualitative description of the environment that is suitable for robotics applications like place recognition ([1], [2], [3]).”
BackgroundMethodOmnidirectional CNN for Visual Place Recognition and Navigation
Wang, T.-H. J., Huang, H.-J., Lin, J.-T., Hu, C.-W., Zeng, K.-H., & Sun, M. (2018) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2018.8463173 · Cited by 94
Experiments in place recognition using gist panoramas
Murillo, A. C., & Košecká, J. (2009) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2009.5457552 · Cited by 78
“In [3] the scalability and speed were achieved by using a version of vocabulary tree along with inverted file index, used to retrieve the top k-views with the closest visual word signatures.”
BackgroundMethodLarge scale place recognition in 2D LIDAR scans using Geometrical Landmark Relations
Himstedt, M., Frost, J., Hellbach, S., Böhme, H.-J., & Maehle, E. (2014) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2014.6943277 · Cited by 97
“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].”
BackgroundMethodVisual place recognition using HMM sequence matching
Hansen, P., & Browning, B. (2014) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2014.6943207 · Cited by 89
“The FAB-MAP algorithm [3], [4] remains a popular state-of-the-art algorithm achieving robust image recall performance for outdoor image sequences up to 1000 kilometers.”
MethodSimultaneous Localization And Mapping: Present, Future, and the Robust-Perception Age.Preprint
Cadena, C., Carlone, L., Carrillo, H., Latif, Y., Scaramuzza, D., Neira, J., Reid, I. D., & Leonard, J. J. (2016) · arXiv (Cornell University) · Cited by 90
The gist of maps - summarizing experience for lifelong localization
Dymczyk, M., Lynen, S., Cieslewski, T., Bosse, M. C., Siegwart, R., & Furgale, P. (2015) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2015.7139575 · Cited by 86
“Although FAB-MAP [19] has been one of the most widely successful place recognition algorithms, its performance degrades when places are visited multiple times [20], because the probability mass gets distributed between the multiple visits if they have not been correctly associated to belonging to the same place.”
BackgroundMemon, A. R., Wang, H., & Hussain, A. (2020) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2020.103470 · Cited by 82
Sensor-level computer vision with pixel processor arrays for agile robots
Dudek, P., Richardson, T. S., Bose, L., Carey, S. J., Chen, J., Greatwood, C. M., Liu, Y., & Mayol-Cuevas, W. W. (2022) · Science Robotics · doi:10.1126/scirobotics.abl7755 · Cited by 87
“(88,89), which via a collection of images can describe a space in a topological manner.”
BackgroundAppearance-invariant place recognition by discriminatively training a convolutional neural network
López-Antequera, M., Gomez-Ojeda, R., Petkov, N., & González-Jiménez, J. (2017) · Pattern Recognition Letters · doi:10.1016/j.patrec.2017.04.017 · Cited by 80
Scalable Place Recognition Under Appearance Change for Autonomous Driving
Doan, A.-D., Latif, Y., Chin, T.-J., Liu, Y., Do, T.-T., & Reid, I. D. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00941 · Cited by 88
“Traditionally, it has been posed as an image retrieval problem using local features aggregated via a BoW representation [10, 13, 11].”
Efficient Non-Consecutive Feature Tracking for Robust Structure-From-Motion
Zhang, G., Liu, H., Dong, Z., Jia, J., Wong, T., & Bao, H. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2607425 · Cited by 77
“Cummins and Newman [26] proposed a probabilistic approach called FAB-MAP for location recognition and online loop closure detection, which models the world as a set of locations and computes the probability of belonging to previously visited locations for each input image.”
BackgroundMethodIndoor Relocalization in Challenging Environments With Dual-Stream Convolutional Neural Networks
Li, R., Liu, Q., Gui, J., Gu, D., & Hu, H. (2017) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2017.2664920 · Cited by 80
“Cummins and Newman [16] presented FAB-MAP with SURF point features as visual words to implement place recognition and relocalization in large-scale environments.”
MethodOzog, P., Carlevaris‐Bianco, N., Kim, A., & Eustice, R. M. (2015) · Journal of Field Robotics · doi:10.1002/rob.21582 · Cited by 82
“Cummins and Newman (2008) used a generative model to describe the co-occurrence of words in the training data.”
MethodMade to measure: Bespoke landmarks for 24-hour, all-weather localisation with a camera
Linegar, C., Churchill, W., & Newman, P. (2016) · 2016 IEEE International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra.2016.7487208 · Cited by 76
“SeqSLAM, by exploiting the sequence information, demonstrated day-to-night localisation, and noted that FAB-MAP [2] (a point-based feature system) performed poorly in such conditions.”
MethodSelf-citationVisual mapping for natural gas pipe inspection
Hansen, P., Alismail, H., Rander, P. W., & Brett Browning, A. (2014) · The International Journal of Robotics Research · doi:10.1177/0278364914550133 · Cited by 75
“Popular algorithms such as FABMAP (Cummins and Newman, 2008, 2011) have evolved from the visual ‘bag of words’ (BoW) method presented in Sivic and Zisserman (2003).”
MethodLoop closure detection for visual SLAM systems using convolutional neural network
Zhang, X., Su, Y. L., & Zhu, X. (2017) · International Conference on Automation and Computing (ICAC) · doi:10.23919/iconac.2017.8082072 · Cited by 79
“Experiments are demonstrated on open datasets from the New College and City Centre sequence [11] to test the feasibility of the presented algorithm.”
MethodInfluentialGarg, S., Sünderhauf, N., & Milford, M. J. (2018) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2018.8461051 · Cited by 80
Large scale multiple robot visual mapping with heterogeneous landmarks in semi-structured terrain
Vidal-Calleja, T. A., Berger, C., Solà, J., & Lacroix, S. (2011) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2011.05.008 · Cited by 71
“image’s descriptors matching (SIFT, SURF), image indexing techniques [26] or scan matching.”
MethodOnline and Incremental Appearance-based SLAM in Highly Dynamic Environments
Kawewong, A., Tongprasit, N., Tangruamsub, S., & Hasegawa, O. (2011) · The International Journal of Robotics Research · doi:10.1177/0278364910371855 · Cited by 70
“We utilized two datasets used earlier (Cummins and Newman 2008a), City Centre and New College, to sufficiently evaluate the PIRF-Nav performance under strong perceptual aliasing and under dynamic scene changes.”
BackgroundMethodResultInfluentialMobile robot vision navigation & localization using Gist and Saliency
Chang, C.-K., Siagian, C., & Itti, L. (2010) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2010.5649136 · Cited by 73
BoRF: Loop-closure detection with scale invariant visual features
Zhang, H. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5980273 · Cited by 70
“Techniques for reducing perceptual aliasing include: increasing the vocabulary size, the use of word correlation [5], rank-reduction [13], discriminative vocabulary selection [18], word-weighting with tf-idf (term-”
BackgroundMethodVisual Place Recognition: A Tutorial [Tutorial]
Schubert, S., Neubert, P., Garg, S., Milford, M. J., & Fischer, T. (2023) · IEEE Robotics & Automation Magazine · doi:10.1109/mra.2023.3310859 · Cited by 102
“This approach re-ranks the top-K retrieved matches from holistic descriptors through geometric verification with local image descriptors [9], [65].”
BackgroundMethodOnline probabilistic topological mapping
Ranganathan, A., & Dellaert, F. (2011) · The International Journal of Robotics Research · doi:10.1177/0278364910393287 · Cited by 68
“Cummins and Newman (2008) also provide an appearance-based technique for loop closing based on generative-modeling of bag-of-words models of location images.”
BackgroundLocalization from semantic observations via the matrix permanent
Atanasov, N., Zhu, M., Daniilidis, K., & Pappas, G. J. (2015) · The International Journal of Robotics Research · doi:10.1177/0278364915596589 · Cited by 80
“The conditional entropy is an appropriate objective because it quantifies the amount of information needed to describe the outcome of a random quantity (the future poses) given the value of another random quantity (the semantic observations) (Cover & Thomas 2006).”
BackgroundBTC: A Binary and Triangle Combined Descriptor for 3-D Place Recognition
Yuan, C., Lin, J., Liu, Z., Wei, H., Hong, X., & Zhang, F. (2024) · IEEE Transactions on Robotics · doi:10.1109/tro.2024.3353076 · Cited by 98
“Oneapproachreliesonimagescaptured byvisualcameras,suchasFAB-MAP[12],DBoW2[13],and SeqSLAM[14].”
MethodDiscovering and exploiting 3D symmetries in structure from motion
Cohen, A., Zach, C., Sinha, S. N., & Pollefeys, M. (2012) · 2012 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2012.6247841 · Cited by 67
S2DNet: Learning Image Features for Accurate Sparse-to-Dense Matching
Germain, H., Bourmaud, G., & Lepetit, V. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58580-8_37 · Cited by 83
Learning to close loops from range data
Granström, K., Schön, T. B., Nieto, J. I., & Ramos, F. T. (2011) · The International Journal of Robotics Research · doi:10.1177/0278364911405086 · Cited by 67
“In particular, methods using vision sensors have broadly been presented (see, e.g., Goedeme et al. 2006; Tapus and Siegwart 2006; Fraundorfer et al. 2007; Ho and Newman 2007; Angeli et al. 2008; Cummins and Newman 2008; Callmer et al. 2008; Eade and Drummond 2008; Milford and Wyeth 2008; Cummins and Newman 2009; Konol…”
BackgroundMethodResultInfluentialUsing superpixels in monocular SLAM
Concha, A., & Civera, J. (2014) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2014.6906883 · Cited by 78
Indoor scene recognition by a mobile robot through adaptive object detection
Espinace, P., Kollar, T., Roy, N., & Soto, Á. (2013) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2013.05.002 · Cited by 74
“In terms of robotics, besides the fact that some of the methods described above are applied to this field, extensive work has been done in the case of topological localization using visual landmarks [5,43].”
Method
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.