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
Variational Inference: A Review for Statisticians
Blei, D. M., Kucukelbir, A., & McAuliffe, J. D. (2017) · Journal of the American Statistical Association · doi:10.1080/01621459.2017.1285773 · Cited by 3,811
“…Wang and Mori 2009), image denoising (Likas and Galatsanos 2004), tracking (Vermaak, Lawrence, and Pérez 2003; Yu and Wu 2005), place recognition and mapping for robotics (Cummins and Newman 2008; Ramos et al. 2012), and image segmentation with Bayesian nonparametrics (Sudderth and Jordan 2009).”
BackgroundPast, 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
“Short-term data association is responsible for associating corresponding features in consecutive sensor measurements; for instance, short-term data association would track the fact that 2 pixel measurements in consecutive frames are picturing the same 3D point.”
NetVLAD: CNN Architecture for Weakly Supervised Place Recognition
Arandjelovic, R., Gronat, P., Torii, A., Pajdla, T., & Sivic, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.572 · Cited by 2,584
“Visual place recognition has received a significant amount of attention in the past years both in computer vision [5, 10, 11, 24, 35, 62, 63, 64, 65, 79, 80] and robotics communities [16, 17, 44, 46, 74] motivated by, e.”
BackgroundPast, 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
PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization
Kendall, A., Grimes, M., & Cipolla, R. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.336 · Cited by 2,489
“Appearance-based relocalization has had success [4, 23] in coarsely locating the camera among a limited, discretized set of place labels, leaving the pose estimation to a separate system.”
BackgroundMethodVariational Inference: A Review for Statisticians
Blei, D. M., Kucukelbir, A., & McAuliffe, J. (2017) · Figshare · doi:10.6084/m9.figshare.5203696 · Cited by 1,919
Bags of Binary Words for Fast Place Recognition in Image Sequences
Galvez-López, D., & Tardos, J. D. (2012) · IEEE Transactions on Robotics · doi:10.1109/tro.2012.2197158 · Cited by 1,913
“An example of this is the FAB-MAP system [13], which detects loops with an omnidirectional camera, obtaining a recall of 48.4% and 3.1%, with no false positives, in trajectories 70 and 1000 km in length.”
BackgroundMethodInfluentialBadue, C., Guidolini, R., Carneiro, R. V., Azevedo, P., Cardoso, V. B., Forechi, A., Jesus, L., Berriel, R., Paixão, T. M., Mutz, F., et al. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2020.113816 · Cited by 1,209
Visual Place Recognition: A Survey
Lowry, S., Sünderhauf, N., Newman, P., Leonard, J. J., Cox, D., Corke, P., & Milford, M. J. (2016) · IEEE Transactions on Robotics · doi:10.1109/tro.2015.2496823 · Cited by 1,079
“FAB-MAP [6], [87] is a probabilistic appearance-based localization system that uses a data-driven approach to calculating an observational likelihood.”
BackgroundMethodInfluentialSelf-citationImage Matching from Handcrafted to Deep Features: A Survey
Ma, J., Jiang, X., Fan, A., Jiang, J., & Yan, J. (2021) · International Journal of Computer Vision · doi:10.1007/s11263-020-01359-2 · Cited by 990
“The presence of outliers has been verified to be the reason of performance degradation for visual homing (Schroeter and Newman 2008).”
MethodScan Context: Egocentric Spatial Descriptor for Place Recognition Within 3D Point Cloud Map
Kim, G., & Kim, A. (2018) · 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8593953 · Cited by 904
“FAB-MAP [13] increased robustness with the probabilistic approach by learning a generative model for the bag of visual words.”
MethodSimultaneous 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
“Algorithms like FAB-MAP (Fast Appearance Based Map) [99] (and its extension FAB-MAP 2.0 [137]) and PIRF-Nav (Position Invariant Robust Feature Navigation) [138] are two examples of SLAM solutions based on landmark appearance for loop closures.”
InfluentialVisual simultaneous localization and mapping: a survey
Fuentes-Pacheco, J., Ruiz-Ascencio, J., & Rendón-Mancha, J. M. (2015) · Artificial Intelligence Review · doi:10.1007/s10462-012-9365-8 · Cited by 883
“Many BoVW schemes generate an off-line vocabulary by means of a K-means clustering (but any other can be used) of descriptors from a large corpus of training images (Ho and Newman 2007; Cummins and Newman 2008).”
BackgroundInfluentialGeometric Loss Functions for Camera Pose Regression with Deep Learning
Kendall, A., & Cipolla, R. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.694 · Cited by 803
“Traditionally, this has been modelled as an image retrieval problem [6, 9, 53, 45] enabling the use of efficient and scalable retrieval approaches [35, 38] such as Bag-of-Words (BoW) [47], VLAD [17, 10], and Fisher vectors [18].”
MethodVisual SLAM algorithms: a survey from 2010 to 2016
Taketomi, T., Uchiyama, H., & Ikeda, S. (2017) · IPSJ Transactions on Computer Vision and Applications · doi:10.1186/s41074-017-0027-2 · Cited by 725
“They used a bag-of-words based image retrieval technique to detect one of keyframes which view is similar with the current view [36].”
MethodBenchmarking 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
PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition
Uy, M. A., & Lee, G. H. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00470 · Cited by 670
Computer Vision for Autonomous Vehicles: Problems, Datasets and State of the Art
Janai, J., Güney, F., Behl, A., & Geiger, A. (2020) · Foundations and Trends® in Computer Graphics and Vision · doi:10.1561/0600000079 · Cited by 643
Robust reconstruction of indoor scenes
Choi, S., Zhou, Q.-Y., & Koltun, V. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7299195 · Cited by 555
“Global optimization of range scan poses based on hypothesized pairwise relations was introduced by Lu and Milios [41] and is commonly used in robotics [13, 33, 23].”
MethodVisual Odometry : Part II: Matching, Robustness, Optimization, and Applications
Fraundorfer, F., & Scaramuzza, D. (2012) · IEEE Robotics & Automation Magazine · doi:10.1109/mra.2012.2182810 · Cited by 620
Appearance-only SLAM at large scale with FAB-MAP 2.0
Cummins, M., & Newman, P. (2011) · The International Journal of Robotics Research · doi:10.1177/0278364910385483 · Cited by 609
“Keywords Robotics, SLAM, appearance-based navigation, FAB-MAP”
BackgroundMethodResultInfluentialSelf-citationPL-SLAM: A Stereo SLAM System Through the Combination of Points and Line Segments
Gomez-Ojeda, R., Moreno, F.-A., Zuniga-Noel, D., Scaramuzza, D., & Gonzalez-Jimenez, J. (2019) · IEEE Transactions on Robotics · doi:10.1109/tro.2019.2899783 · Cited by 560
University of Michigan North Campus long-term vision and lidar dataset
Carlevaris-Bianco, N., Ushani, A. K., & Eustice, R. M. (2016) · The International Journal of Robotics Research · doi:10.1177/0278364915614638 · Cited by 576
“Though the same area is repeatedly explored, the path for each session is varied, as is the time of day for each session, from early morning to just after dusk.”
BackgroundModelling uncertainty in deep learning for camera relocalization
Kendall, A., & Cipolla, R. (2016) · 2016 IEEE International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra.2016.7487679 · Cited by 548
“This has been achieved using SIFT [2] features with FAB-MAP [6] or using convolutional neural network features [7].”
MethodOn the performance of ConvNet features for place recognition
Sunderhauf, N., Shirazi, S., Dayoub, F., Upcroft, B., & Milford, M. (2015) · 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2015.7353986 · Cited by 544
“Previous work [3], [32] established that FAB-MAP [18], [19] is not capable of handling the severe appearance changes of the Nordland and St. Lucia datasets.”
BackgroundGeometry-Aware Learning of Maps for Camera Localization
Brahmbhatt, S., Gu, J., Kim, K., Hays, J., & Kautz, J. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00277 · Cited by 530
Fast image-based localization using direct 2D-to-3D matching
Sattler, T., Leibe, B., & Kobbelt, L. (2011) · 2011 International Conference on Computer Vision · doi:10.1109/iccv.2011.6126302 · Cited by 480
“Modern SLAM methods enable this approach to scale to larger scenes, while still permitting real-time localization and mapping [3, 5, 6].”
BackgroundPatch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition
Hausler, S., Garg, S., Xu, M., Milford, M., & Fischer, T. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.01392 · Cited by 444
“Patch-NetVLAD won the Mapillary Challenge at the ECCV2020 workshops (not yet publicly announced to comply with CVPR’s double-blind policy), with Table 1 showing that Patch-NetVLAD outperformed the baseline method, NetVLAD, by 13.0% (absolute R@1 increase) on the withheld test dataset.”
MethodInfluentialThe Oxford Radar RobotCar Dataset: A Radar Extension to the Oxford RobotCar Dataset
Barnes, D., Gadd, M., Murcutt, P., Newman, P., & Posner, I. (2019) · arXiv · doi:10.48550/arxiv.1909.01300 · Cited by 443
“The trajectories were generated by optimising robust VO [20], visual loop closures [21], and GPS/INS as constraints.”
MethodSelf-citationThe Oxford Radar RobotCar Dataset: A Radar Extension to the Oxford RobotCar Dataset
Barnes, D., Gadd, M., Murcutt, P., Newman, P., & Posner, I. (2020) · 2020 IEEE International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra40945.2020.9196884 · Cited by 429
“The trajectories were generated by optimising robust VO [20], visual loop closures [21], and GPS/INS as constraints.”
MethodSelf-citationCity-scale landmark identification on mobile devices
Chen, D. M., Baatz, G., Koser, K., Tsai, S. S., Vedantham, R., Pylvanainen, T., Roimela, K., Chen, X., Bach, J., Pollefeys, M., et al. (2011) · CVPR 2011 · doi:10.1109/cvpr.2011.5995610 · Cited by 400
“Cummins and Newman [4] propose a probabilistic approach to the problem of recognizing places based on co-occurences of visual words and learn a generative model of place appearance .”
BackgroundInfluentialA survey of state-of-the-art on visual SLAM
Abaspur Kazerouni, I., Fitzgerald, L., Dooly, G., & Toal, D. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.117734 · Cited by 383
A Review of Visual-LiDAR Fusion based Simultaneous Localization and Mapping
Debeunne, C., & Vivet, D. (2020) · Sensors · doi:10.3390/s20072068 · Cited by 386
Visual SLAM and Structure from Motion in Dynamic Environments: A Survey
Saputra, M. R. U., Markham, A., & Trigoni, N. (2019) · ACM Computing Surveys · doi:10.1145/3177853 · Cited by 398
“…like ORB-SLAM [113] integrate hardware and algorithmic advancement in the past decade by including parallel computing, ORB features [131], statistical model selection [155], loop closures based on bag-of-words place recognition [26, 41], local bundle adjustment [111], and graph optimization [81].”
MethodPlace Recognition with ConvNet Landmarks: Viewpoint-Robust, Condition-Robust, Training-Free
Suenderhauf, N., Shirazi, S., Jacobson, A., Dayoub, F., Pepperell, E., Upcroft, B., & Milford, M. (2015) · Robotics: Science and Systems XI · doi:10.15607/rss.2015.xi.022 · Cited by 379
“Concretely, we compare against the whole image, single ConvNet feature system proposed by [41], the featurebased method FAB-MAP [7] (using OpenFABMAP [14]) and the sequence-based approaches SeqSLAM [27] (using the”
BackgroundMethodResultInfluentialAppearance-Based Loop Closure Detection for Online Large-Scale and Long-Term Operation
Labbe, M., & Michaud, F. (2013) · IEEE Transactions on Robotics · doi:10.1109/tro.2013.2242375 · Cited by 372
“Not merging locations helps to keep different signatures of the same location for better hypothesis estimation, which is important in a highly dynamic environment or when the environment changes gradually over time in a cyclic way (e.g., day-night or weather variations).”
BackgroundDeep learning features at scale for visual place recognition
Chen, Z., Jacobson, A., Sunderhauf, N., Upcroft, B., Liu, L., Shen, C., Reid, I., & Milford, M. (2017) · 2017 IEEE International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra.2017.7989366 · Cited by 344
VINet: Visual-Inertial Odometry as a Sequence-to-Sequence Learning Problem
Clark, R., Wang, S., Wen, H., Markham, A., & Trigoni, N. (2017) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v31i1.11215 · Cited by 327
“FAB-MAP (Cummins and Newman 2008)) are integrated to reduce scale drift.”
BackgroundLDSO: Direct Sparse Odometry with Loop Closure
Gao, X., Wang, R., Demmel, N., & Cremers, D. (2018) · 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8593376 · Cited by 327
MulRan: Multimodal Range Dataset for Urban Place Recognition
Kim, G., Park, Y. S., Cho, Y., Jeong, J., & Kim, A. (2020) · 2020 IEEE International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra40945.2020.9197298 · Cited by 301
“For instance, FAB-MAP [11] solved for a large scale appearance SLAM via probabilistic formulation.”
MethodRobust map optimization using dynamic covariance scaling
Agarwal, P., Tipaldi, G. D., Spinello, L., Stachniss, C., & Burgard, W. (2013) · 2013 IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2013.6630557 · Cited by 293
“Generating outlier-free graphs in the front-end, however, is very challenging, especially in environments showing selfsimilar structures [5], [6], [7].”
BackgroundTotal recall II: Query expansion revisited
Chum, O., Mikulik, A., Perdoch, M., & Matas, J. (2011) · CVPR 2011 · doi:10.1109/cvpr.2011.5995601 · Cited by 287
“However, it has been observed in [8, 6, 13] that the assumption about noisy features does not always hold.”
BackgroundPersistent Navigation and Mapping using a Biologically Inspired SLAM System
Milford, M., & Wyeth, G. (2010) · The International Journal of Robotics Research · doi:10.1177/0278364909340592 · Cited by 286
“As a primarily appearance-based SLAM system, it is one of a number of recent appearance-based systems (Andreasson et al., 2008 Angeli et al., 2008 Cummins and Newman, 2008 Mahon et al., 2008) that have been catching up to large-scale metric mapping systems.”
BackgroundHighly scalable appearance-only SLAM - FAB-MAP 2.0
Cummins, M., & Newman, P. (2009) · Robotics: Science and Systems V · doi:10.15607/rss.2009.v.039 · Cited by 282
“In comparison to prior work [1], the new system’s inference”
BackgroundMethodResultInfluentialSelf-citationA comparison of loop closing techniques in monocular SLAM
Williams, B., Cummins, M., Neira, J., Newman, P., Reid, I., & Tardós, J. (2009) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2009.06.010 · Cited by 283
Self-citationSchonberger, J. L., Pollefeys, M., Geiger, A., & Sattler, T. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00721 · Cited by 267
Improving Image-Based Localization by Active Correspondence Search
Sattler, T., Leibe, B., & Kobbelt, L. (2012) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-33718-5_54 · Cited by 261
“It is encountered in many interesting applications, such as touristic landmark recognition [1, 2], robot localization and loop closure [3, 4], Augmented Reality [5, 6], and Structure-from-Motion [7].”
BackgroundMethodReal-Time Visual SLAM for Autonomous Underwater Hull Inspection Using Visual Saliency
Kim, A., & Eustice, R. M. (2013) · IEEE Transactions on Robotics · doi:10.1109/tro.2012.2235699 · Cited by 263
“This is in contrast with visual place recognition techniques, which typically have vocabulary sizes in the 4000– 11 000 range or more [50], [55]–[58].”
BackgroundMethodResultInfluentialVisual teach and repeat for long‐range rover autonomy
Furgale, P., & Barfoot, T. D. (2010) · Journal of Field Robotics · doi:10.1002/rob.20342 · Cited by 266
The Málaga urban dataset: High-rate stereo and LiDAR in a realistic urban scenario
Blanco-Claraco, J.-L., Moreno-Dueñas, F.-Á., & González-Jiménez, J. (2014) · The International Journal of Robotics Research · doi:10.1177/0278364913507326 · Cited by 262
“: color 640 × 480@ ∼ 1 fps College: 2 Oxford city center City: 28 (Cummins and Newman, 2008) New College (2009) × 2 Stereo: b/w 512 × 384@20 fps 2.”
Fast relocalisation and loop closing in keyframe-based SLAM
Mur-Artal, R., & Tardos, J. D. (2014) · 2014 IEEE International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra.2014.6906953 · Cited by 264
“They use randomized list of binary test classifiers to find correspondences between image features and previously trained map features.”
Visual place recognition: A survey from deep learning perspective
Zhang, X., Wang, L., & Su, Y. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2020.107760 · Cited by 261
Deep Auxiliary Learning for Visual Localization and Odometry
Valada, A., Radwan, N., & Burgard, W. (2018) · 2018 IEEE International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra.2018.8462979 · Cited by 241
LCDNet: Deep Loop Closure Detection and Point Cloud Registration for LiDAR SLAM
Cattaneo, D., Vaghi, M., & Valada, A. (2022) · IEEE Transactions on Robotics · doi:10.1109/tro.2022.3150683 · Cited by 258
“Traditionally, vision-based techniques for loop closure detection rely on handcrafted features for identifying and representing relevant parts of scenes depicted within images, and exploit a Bag-of-Words model to combine them [5], [6].”
BackgroundGet Out of My Lab: Large-scale, Real-Time Visual-Inertial Localization
Lynen, S., Sattler, T., Bosse, M., Hesch, J., Pollefeys, M., & Siegwart, R. (2015) · Robotics: Science and Systems XI · doi:10.15607/rss.2015.xi.037 · Cited by 243
“In order to match the features against the 3D landmarks, approaches from robotics can be used to first retrieve relevant database images [8, 16, 48].”
MethodComputer Vision for Autonomous Vehicles: Problems, Datasets and State of the Art
Janai, J., Güney, F., Behl, A., & Geiger, A. (2020) · doi:10.1561/9781680836899 · Cited by 232
Are We There Yet? Challenging SeqSLAM on a 3000 km Journey Across All Four Seasons
Suenderhauf, N., Neubert, P., & Protzel, P. (2013) · QUT ePrints (Queensland University of Technology) · Cited by 241
A survey on Visual-Based Localization: On the benefit of heterogeneous data
Piasco, N., Sidibé, D., Demonceaux, C., & Gouet-Brunet, V. (2018) · Pattern Recognition · doi:10.1016/j.patcog.2017.09.013 · Cited by 236
“SURF descriptors [18], light version of SIFT, are employed when real-time performance are required [44, 158, 191].”
BackgroundMethodNight-to-Day Image Translation for Retrieval-based Localization
Anoosheh, A., Sattler, T., Timofte, R., Pollefeys, M., & Gool, L. V. (2019) · 2019 International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra.2019.8794387 · Cited by 224
“These include structure-based localization techniques ActiveSearch [22] and CSL [24], in addition to the image-based FAB-MAP [6] and NetVLAD [2].”
MethodConvolutional Neural Network-based Place Recognition
Chen, Z., Lam, O., Jacobson, A., & Milford, M. J. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1411.1509 · Cited by 201
“State-of-the-art SLAM systems such as FAB-MAP [Cummins and Newman, 2008] further represent appearance data using sets of local features, converting images into “bag-of-words”, which enables efficient retrieval.”
MethodFAB-MAP + RatSLAM: Appearance-based SLAM for multiple times of day
Glover, A. J., Maddern, W. P., Milford, M. J., & Wyeth, G. F. (2010) · 2010 IEEE International Conference on Robotics and Automation · doi:10.1109/robot.2010.5509547 · Cited by 213
“periments [10] in which isual location is created new location is > 0.”
BackgroundMethodInfluentialTowards a robust back-end for pose graph SLAM
Sunderhauf, N., & Protzel, P. (2012) · 2012 IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2012.6224709 · Cited by 198
“Furthermore, the front-end part of the system can detect loop closures between two poses xi and xj , for example by a visual place recognition system like [4] or [29].”
BackgroundLong-Term Visual Localization Revisited
Toft, C., Maddern, W., Torii, A., Hammarstrand, L., Stenborg, E., Safari, D., Okutomi, M., Pollefeys, M., Sivic, J., Pajdla, T., et al. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3032010 · Cited by 190
“Landmarks for Place Recognition [95] 0.0 / 0.2 / 20.8 0.0 / 1.0 / 10.2 FAB-MAP [21] 0.0 / 0.0 / 4.6 0.0 / 0.0 / 0.0 It is interesting to note that while the hierarchical methods experience a large drop for night-time images in the high-precision regime, the drop in accuracy for the coarse-precision…”
BackgroundMethodInfluentialA Survey on Deep Visual Place Recognition
Masone, C., & Caputo, B. (2021) · IEEE Access · doi:10.1109/access.2021.3054937 · Cited by 192
“Although FAB-MAP can work as a pure image retrieval process by always assuming a uniform location prior, using even a simple motion model to leverage the prior estimate improves performance.”
MethodInfluentialRSLAM: A System for Large-Scale Mapping in Constant-Time Using Stereo
Mei, C., Sibley, G., Cummins, M., Newman, P., & Reid, I. (2011) · International Journal of Computer Vision · doi:10.1007/s11263-010-0361-7 · Cited by 193
“For loop closure we rely on fast appearance based mapping (Cummins and Newman 2008).”
Self-citationAutonomous MAV flight in indoor environments using single image perspective cues
Bills, C., Chen, J., & Saxena, A. (2011) · 2011 IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2011.5980136 · Cited by 193
“, [26], [27], [28]), in which the map is built using images captured from a camera.”
BackgroundBRIEF-Gist - closing the loop by simple means
Sunderhauf, N., & Protzel, P. (2011) · 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems · doi:10.1109/iros.2011.6094921 · Cited by 184
“The dataset features several loop closings in a dynamic environment including moving people, as well as different types of environment (urban, park) and changing lighting conditions.”
BackgroundInfluentialTransVPR: Transformer-Based Place Recognition with Multi-Level Attention Aggregation
Wang, R., Shen, Y., Zuo, W., Zhou, S., & Zheng, N. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01328 · Cited by 185
“In early VPR systems [1, 11, 25, 30, 31, 46], traditional methods such as SIFT [26], SURF [4] and ORB [39] have been widely used to represent a small patch centered around a detected keypoint.”
MethodFast LIDAR localization using multiresolution Gaussian mixture maps
Wolcott, R. W., & Eustice, R. M. (2015) · 2015 IEEE International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra.2015.7139582 · Cited by 188
“While several groups have attempted to transition to vision-only solutions for autonomous vehicles because of cost and visual appearance [1]–[4], manufacturers continue to lower the price and increase the aesthetic appeal of 3D LIDAR scanners [5].”
BackgroundUnsupervised learning to detect loops using deep neural networks for visual SLAM system
Gao, X., & Zhang, T. (2017) · Autonomous Robots · doi:10.1007/s10514-015-9516-2 · Cited by 186
“To further evaluate our method, we perform another experiment using the “New College” and “City Centre” sequence provided by Fabmap.”
MethodInfluentialInformation-Based Compact Pose SLAM
Ila, V., Porta, J. M., & Andrade-Cetto, J. (2010) · IEEE Transactions on Robotics · doi:10.1109/tro.2009.2034435 · Cited by 183
“Our Pose-SLAM method was compared with a state-of-the-art placerecognition technique [39].”
MethodX-View: Graph-Based Semantic Multi-View Localization
Gawel, A., Don, C. D., Siegwart, R., Nieto, J., & Cadena, C. (2018) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2018.2801879 · Cited by 183
Vision-based topological mapping and localization methods: A survey
Garcia-Fidalgo, E., & Ortiz, A. (2015) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2014.11.009 · Cited by 175
3D model matching with Viewpoint-Invariant Patches (VIP)
Wu, C., Clipp, B., Li, X., Frahm, J.-M., & Pollefeys, M. (2008) · 2008 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2008.4587501 · Cited by 171
Tsintotas, K. A., Bampis, L., & Gasteratos, A. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2022.3175656 · Cited by 175
“5 km X City Centre [113] Stereo, color, lateral Outdoor, urban, dynamic 1024 × 768, 7 Hz 1237 ∼1.”
BackgroundMethodInfluentialConvolutional neural network-based image representation for visual loop closure detection
Hou, Y., Zhang, H., & Zhou, S. (2015) · 2015 IEEE International Conference on Information and Automation · doi:10.1109/icinfa.2015.7279659 · Cited by 171
Towards semantic SLAM using a monocular camera
Civera, J., Galvez-Lopez, D., Riazuelo, L., Tardos, J. D., & Montiel, J. M. M. (2011) · 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems · doi:10.1109/iros.2011.6094648 · Cited by 172
“Visual recognition has been used in robotic SLAM mostly for scene recognition, for example loop closing in [7], [2], relocalization in [28] and place classification in a semantic context in [22].”
MethodPlace recognition using keypoint voting in large 3D lidar datasets
Bosse, M., & Zlot, R. (2013) · 2013 IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2013.6630945 · Cited by 171
“One popular approach is the bag-of-words (BoW) technique [4], [16], [7], in which the place descriptors are vectors containing binary values or weighted histograms accumulating occurrences of keypoint descriptors quantized to words in a learned vocabulary.”
iBoW-LCD: An Appearance-Based Loop-Closure Detection Approach Using Incremental Bags of Binary Words
Garcia-Fidalgo, E., & Ortiz, A. (2018) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2018.2849609 · Cited by 167
“information to undertake the association problem is generically known as appearance-based loop closure detection [2]–[9].”
BackgroundMethodInfluentialDXSLAM: A Robust and Efficient Visual SLAM System with Deep Features
Li, D., Shi, X., Long, Q., Liu, S., Yang, W., Wang, F., Wei, Q., & Qiao, F. (2020) · 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9340907 · Cited by 154
“One of the most successful applications of this method is FAB-MAP, which employs SURF and BoW to train vocabulary for place recognition and demonstrates robust performance against viewpoint changes [31].”
MethodInfluentialSPP-Net: Deep Absolute Pose Regression with Synthetic ViewsPreprint
Purkait, P., Zhao, C. B., & Zach, C. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1712.03452 · Cited by 238
OpenFABMAP: An open source toolbox for appearance-based loop closure detection
Glover, A., Maddern, W., Warren, M., Reid, S., Milford, M., & Wyeth, G. (2012) · 2012 IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2012.6224843 · Cited by 162
“A more in depth description can be found in [6] and [8].”
BackgroundVisual SLAM for autonomous ground vehicles
Lategahn, H., Geiger, A., & Kitt, B. (2011) · 2011 IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2011.5979711 · Cited by 159
“It can be seen that the accuracy of the filtered point is considerably better.”
BackgroundOptimal Transport Aggregation for Visual Place Recognition
Izquierdo, S., & Civera, J. (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/cvpr52733.2024.01672 · Cited by 330
“Early approaches to VPR used either aggregations of handcrafted local features [3, 15, 26] or global descriptors [39, 53].”
MethodVast-scale Outdoor Navigation Using Adaptive Relative Bundle Adjustment
Sibley, G., Mei, C., Reid, I., & Newman, P. (2010) · The International Journal of Robotics Research · doi:10.1177/0278364910369268 · Cited by 156
“For loop closure and place recognition we rely on FAB-MAP, which represents each place using the bag-of-words model developed for image retrieval systems by the computer vision community Sivic and Zisserman (2006) Cummins and Newman (2008).”
BackgroundMethodSelf-citationKhaliq, A., Ehsan, S., Chen, Z., Milford, M. J., & McDonald-Maier, K. D. (2019) · IEEE Transactions on Robotics · doi:10.1109/tro.2019.2956352 · Cited by 182
“Against our approach, RMAC on AlexNet365 achieves comparable and better performance than FABMAP and pooling techniques including Sum-Pool, Max-Pool, and Cross-Pool.”
MethodInfluentialExploiting uncertainty in regression forests for accurate camera relocalization
Valentin, J., Niebner, M., Shotton, J., Fitzgibbon, A., Izadi, S., & Torr, P. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7299069 · Cited by 156
“A few successful approaches also exist that do not belong to either of those two classes [5, 16].”
BackgroundOnly look once, mining distinctive landmarks from ConvNet for visual place recognition
Chen, Z., Maffra, F., Sa, I., & Chli, M. (2017) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2017.8202131 · Cited by 172
“Traditional approaches either operate directly on raw pixels [12] or utilize a fixed set of handcrafted features [1] (i.”
BackgroundMethodGlobal Localization from Monocular SLAM on a Mobile Phone
Ventura, J., Arth, C., Reitmayr, G., & Schmalstieg, D. (2014) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2014.27 · Cited by 172
“For example, Cummins and Newman describe appearance-only SLAM, which builds a topograhic map based on visual loop closures [8, 9].”
BackgroundLightweight Unsupervised Deep Loop Closure
Merrill, N., & Huang, G. P. (2018) · Robotics: Science and Systems · doi:10.15607/rss.2018.xiv.032 · Cited by 171
Lost! Leveraging the Crowd for Probabilistic Visual Self-Localization
Brubaker, M. A., Geiger, A., & Urtasun, R. (2013) · 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.2013.393 · Cited by 152
“Furthermore, recent progress in loop-closure detection [4] [21] [26] has led to improved maps by constraining the problem at places which have been visited multiple times.”
OpenRatSLAM: an open source brain-based SLAM system
Ball, D., Heath, S., Wiles, J., Wyeth, G. F., Corke, P., & Milford, M. J. (2013) · Autonomous Robots · doi:10.1007/s10514-012-9317-9 · Cited by 152
“We are also investigating the feasibility of an autonomous or supervised process for tuning the key system parameters based on a user providing a representative dataset, similar in overall process to (Fast Appearance Based Mapping) FAB-MAP codebook generation (Cummins and Newman 2008).”
MethodSelf-Supervised Visual Descriptor Learning for Dense Correspondence
Schmidt, T., Newcombe, R., & Fox, D. (2017) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2016.2634089 · Cited by 148
“Existing SLAM systems that utilize feature descriptors have traditionally used hand-engineered features, such as ORB-SLAM [16], FabMap [5], and BundleFusion [6].”
MethodDeep learning in robotics: survey on model structures and training strategies
Karoly, A. I., Galambos, P., Kuti, J., & Rudas, I. J. (2021) · IEEE Transactions on Systems, Man, and Cybernetics: Systems · doi:10.1109/tsmc.2020.3018325 · Cited by 146
“The results show that the method can output more robust recognition of previously seen scenes than the FAB-MAP method [114], which was used as a baseline for the experiments.”
MethodLarge-Scale Location Recognition and the Geometric Burstiness Problem
Sattler, T., Havlena, M., Schindler, K., & Pollefeys, M. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.175 · Cited by 146
“Location recognition techniques play an important role for several applications such as loop-closure in robotics [9], landmark recognition [4, 7], visual navigation [26], and image-based localization [6,14,29].”
BackgroundZubizarreta, J., Aguinaga, I., & Montiel, J. M. M. (2020) · IEEE Transactions on Robotics · doi:10.1109/tro.2020.2991614 · Cited by 164
“Loop closures are detected using FAB-MAP [13], an appearance loop detection algorithm, which uses different features to those of the direct odometry.”
MethodA Survey on Global LiDAR Localization: Challenges, Advances and Open Problems
Yin, H., Xu, X., Lu, S., Chen, X., Xiong, R., Shen, S., Stachniss, C., & Wang, Y. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02019-5 · Cited by 203
“The filtering technique of FAB-MAP family could handle sequential measurements, but will easily crash when single-shot place recognition fails in challenging scenes.”
InfluentialCloud-Based Collaborative 3D Mapping in Real-Time With Low-Cost Robots
Mohanarajah, G., Usenko, V., Singh, M., D’Andrea, R., & Waibel, M. (2015) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2015.2408456 · Cited by 155
Semantic Graph Based Place Recognition for 3D Point Clouds
Kong, X., Yang, X., Zhai, G., Zhao, X., Zeng, X., Wang, M., Liu, Y., Li, W., & Wen, F. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341060 · Cited by 140
“Research on vision-based place recognition has been investigated for a long time and many successful approaches have been proposed [1]–[3].”
BackgroundMethodReal-time loop detection with bags of binary words
Gálvez‐López, D., & Tardós, J. D. (2011) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2011.6094885 · Cited by 158
“In [4], we also showed that other techniques as FAB-MAP [24] did not reach 100% precision on this dataset.”
MethodInfluential
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.