The DOI matches its Crossref record; 20 of the 21 references it cites were found in a registry or an index.
ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras
Open accessAbstract
We present ORB-SLAM2, a complete simultaneous localization and mapping (SLAM) system for monocular, stereo and RGB-D cameras, including map reuse, loop closing, and relocalization capabilities. The system works in real time on standard central processing units in a wide variety of environments from small hand-held indoors sequences, to drones flying in industrial environments and cars driving around a city. Our back-end, based on bundle adjustment with monocular and stereo observations, allows for accurate trajectory estimation with metric scale. Our system includes a lightweight localization mode that leverages visual odometry tracks for unmapped regions and matches with map points that allow for zero-drift localization. The evaluation on 29 popular public sequences shows that our method achieves state-of-the-art accuracy, being in most cases the most accurate SLAM solution. We publish the source code, not only for the benefit of the SLAM community, but with the aim of being an out-of-the-box SLAM solution for researchers in other fields.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“As an input-driven system, the stochastic input sequences of edge-assisted SLAM cause a considerable impact on the state dynamics and rewards, which then incur high variance during the training of policy gradient methods [13], [14].”
From Orchestrating Joint Offloading and Scheduling for Low-Latency Edge SLAM (2025) · cites this paper as [14]
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“In such an input-driven system, traditional state-dependent DRL algorithms suffer from high state dynamics, leading to inherent bias in policy gradient estimates and increased gradient variance during training [13], [14].”
From Orchestrating Joint Offloading and Scheduling for Low-Latency Edge SLAM (2025) · cites this paper as [14]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ORB: An efficient alternative to SIFT or SURF
Rublee, E., Rabaud, V., Konolige, K., & Bradski, G. (2011) · 2011 International Conference on Computer Vision · doi:10.1109/iccv.2011.6126544 · Cited by 11,037 · cited 1×
“The system uses the same ORB features [17] for tracking, mapping, and place recognition tasks.”
MethodVision meets robotics: The KITTI dataset
Geiger, A., Lenz, P., Stiller, C., & Urtasun, R. (2013) · The International Journal of Robotics Research · doi:10.1177/0278364913491297 · Cited by 9,913 · cited 3×
“We use two different metrics, the absolute translation root-mean-square error (RMSE) tabs proposed in [3], and the average relative translation trel and rotation rrel errors proposed in [2].”
BackgroundMethodORB-SLAM: A Versatile and Accurate Monocular SLAM System
Mur-Artal, R., Montiel, J. M. M., & Tardos, J. D. (2015) · IEEE Transactions on Robotics · doi:10.1109/tro.2015.2463671 · Cited by 6,778 · cited 7×
“Left: Monocular ORB-SLAM [1], right: ORB-SLAM2 (stereo).”
BackgroundMethodResultInfluentialSelf-citationA benchmark for the evaluation of RGB-D SLAM systems
Sturm, J., Engelhard, N., Endres, F., Burgard, W., & Cremers, D. (2012) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2012.6385773 · Cited by 4,230 · cited 4×
“The RGB-D case shows the keyframe poses estimated in sequence fr1_room from the TUM RGB-D Dataset [3], and a dense pointcloud, rendered by backprojecting sensor depth maps from the estimated keyframe poses.”
MethodInfluentialKinectFusion: Real-time dense surface mapping and tracking
Newcombe, R. A., Fitzgibbon, A., Izadi, S., Hilliges, O., Molyneaux, D., Kim, D., Davison, A. J., Kohi, P., Shotton, J., & Hodges, S. (2011) · 2011 10th IEEE International Symposium on Mixed and Augmented Reality · doi:10.1109/ismar.2011.6092378 · Cited by 4,008 · cited 2×
“Note that our SLAM does not perform any fusion like KinectFusion [4] or similar, but the good definition indicates the accuracy of the keyframe poses.”
BackgroundThe EuRoC micro aerial vehicle datasets
Burri, M., Nikolic, J., Gohl, P., Schneider, T., Rehder, J., Omari, S., Achtelik, M., & Siegwart, R. (2016) · The International Journal of Robotics Research · doi:10.1177/0278364915620033 · Cited by 2,086 · cited 1×
“The recent EuRoC dataset [21] contains 11 stereo sequences recorded from a microaerial vehicle (MAV) flying around two different rooms and a large industrial environment.”
BackgroundBags of Binary Words for Fast Place Recognition in Image Sequences
Gálvez‐López, D., & Tardós, J. D. (2012) · IEEE Transactions on Robotics · doi:10.1109/tro.2012.2197158 · Cited by 1,890 · cited 2×
“The system has embedded a Place Recognition module based on DBoW2 [16] for relocalization, in case of tracking failure (e.g. an occlusion) or for reinitialization in an already mapped scene, and for loop detection.”
Dense visual SLAM for RGB-D cameras
Kerl, C., Sturm, J., & Cremers, D. (2013) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2013.6696650 · Cited by 941 · cited 5×
“In Table III we compare our accuracy to the following state-of-the-art methods: ElasticFusion [15], Kintinuous [12], DVO-SLAM [14] and RGB-D SLAM [13].”
MethodInfluential3-D Mapping With an RGB-D Camera
Endres, F., Hess, J., Sturm, J., Cremers, D., & Burgard, W. (2014) · IEEE Transactions on robotics · doi:10.1109/tro.2013.2279412 · Cited by 821 · cited 5×
“In Table III we compare our accuracy to the following state-of-the-art methods: ElasticFusion [15], Kintinuous [12], DVO-SLAM [14] and RGB-D SLAM [13].”
MethodInfluentialVisual-Inertial Monocular SLAM With Map Reuse
Mur-Artal, R., & Tardós, J. D. (2017) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2017.2653359 · Cited by 790 · cited 1×
“As shown in [22], this sequence can be processed using IMU information.”
MethodSelf-citation
Research works printed but not linked: 2
- 10.1177/0278364914551008 · DOI
- kuemmerle, 0, g2o: A general framework for graph optimization, Proc IEEE Int Conf Robot Autom, 3607
Cited by
Sorted by citations
Items 201–300 of 1,004
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. D., Gould, S. J., et al. (2020) · Foundations and Trends in Robotics · doi:10.1561/2300000059 · Cited by 122
RGB-D Inertial Odometry for a Resource-Restricted Robot in Dynamic Environments
Liu, J., Li, X., Liu, Y., & Chen, H. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3191193 · Cited by 110
“[24], state-of-the-art semantic algorithms based on ORB-SLAM2 [4], are also included for comparison.”
Visual SLAM Integration With Semantic Segmentation and Deep Learning: A Review
Pu, H., Luo, J., Wang, G., Huang, T., Liu, H., & Luo, J. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3306371 · Cited by 122
“This approach was inspired by [25] and [106], which involve the semantic segmentation of keyframes in a separate thread and the clustering of depth map using a multiview geometry method.”
BackgroundMethodInfluentialTowards Generalization in Target-Driven Visual Navigation by Using Deep Reinforcement Learning
Devo, A., Mezzetti, G., Costante, G., Fravolini, M. L., & Valigi, P. (2020) · IEEE Transactions on Robotics · doi:10.1109/tro.2020.2994002 · Cited by 114
“For example, it is possible to couple a map-based navigation algorithm [1]– [4] or simultaneous localization and mapping (SLAM) systems [5]–[8] with one of the state-of-the-art object detection or image recognition models [9], [10].”
BackgroundTowards autonomous mapping in agriculture: A review of supportive technologies for ground robotics
Fasiolo, D. T., Scalera, L., Maset, E., & Gasparetto, A. (2023) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2023.104514 · Cited by 121
Beyond Tracking: Selecting Memory and Refining Poses for Deep Visual Odometry
Xue, F., Wang, X., Li, S., Wang, Q., Wang, J., & Zha, H. (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.2019.00877 · Cited by 109
“The results of VISO2-M [10], ORB-SLAM2 [20] (with and without loop closure), and our method are shown in Table 2.”
BackgroundMethodInfluentialMoreFusion: Multi-object Reasoning for 6D Pose Estimation from Volumetric Fusion
Wada, K., Sucar, E., James, S., Lenton, D., & Davison, A. J. (2020) · 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/cvpr42600.2020.01455 · Cited by 107
Kwon, W., Park, J. H., Lee, M., Her, J., Kim, S.-H., & Seo, J.-W. (2019) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2019.2955003 · Cited by 106
“To minimize the disturbance effects caused by the laser scanner’s Hector-SLAM – usually occurring during vertical movement, we add a visual SLAM algorithm (the state-of-the-art is ORB-SLAM2 [12]) implemented by using the upward camera capturing images toward the ceiling.”
MethodInfluentialAugmented Reality Guided Laparoscopic Surgery of the Uterus
Collins, T., Pizarro, D., Gasparini, S., Bourdel, N., Chauvet, P., Canis, M. J., Calvet, L., & Bartoli, A. (2020) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2020.3027442 · Cited by 112
“This experiment is an update of the one presented in [28] with an evaluation of a state-of-the-art monocular SLAM approach (ORB-SLAM) [38].”
BackgroundMethodAccurate Dynamic SLAM Using CRF-Based Long-Term Consistency
Du, Z., Huang, S.-S., Mu, T., Zhao, Q., Martin, R. R., & Xu, K. (2020) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2020.3028218 · Cited by 106
“Local mapping and loop closing are performed as in ORB-SLAM2 [18].”
MethodInfluentialRGB-D SLAM with Structural Regularities
Li, Y., Yunus, R., Brasch, N., Navab, N., & Tombari, F. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561560 · Cited by 104
“Similar to [29], [30], [31], we also use keyframes to build a local map, although our map has point, line and plane landmarks, which are projected into the current frame to search for matches.”
MethodInfluentialUltra-wideband aided fast localization and mapping system
Wang, C., Zhang, H., Nguyen, T.-M., & Xie, L. (2017) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2017.8205968 · Cited by 102
“While the loop closure detection of ORB-SLAM is still enabled to make sure the previous estimation can be corrected when a new visual loop closure is detected.”
MethodInfluentialRobust 2D Indoor Localization Through Laser SLAM and Visual SLAM Fusion
Chan, S.-H., Wu, P.-T., & Fu, L. (2018) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2018.00221 · Cited by 101
“We implement Gmapping algorithm [2] and ORB-SLAM2 algorithm [29] as the choice of laser-based SLAM and visual SLAM.”
MethodLocal feature matching using deep learning: A survey
Xu, S., Chen, S., Xu, R., Wang, C., Lu, P., & Guo, L. (2024) · Information Fusion · doi:10.1016/j.inffus.2024.102344 · Cited by 107
“…foundation for many computer vision tasks, including image fusion [1, 2, 3, 4, 5], visual localization [6, 7, 8, 9], Structure from Motion (SfM) [10, 11, 12, 13], Simultaneous Localization and Mapping (SLAM) [14, 15, 16], optical flow estimation [17, 18, 19], image retrieval [20, 21, 22], and more.”
BackgroundFan, Y., Feng, Z., Mannan, A., Khan, T. U., Shen, C., & Saeed, S. (2018) · Remote Sensing · doi:10.3390/rs10111845 · Cited by 101
“This sensor overcomes the shortcomings of the inability of a monocular SLAM system to obtain the scale and correspondence of a stereo SLAM [30].”
MethodRTG-SLAM: Real-time 3D Reconstruction at Scale using Gaussian Splatting
Peng, Z., Shao, T., Liu, Y., Zhou, J., Yang, Y., Wang, J., & Zhou, K. (2024) · ACM SIGGRAPH Conference (SIGGRAPH) · doi:10.1145/3641519.3657455 · Cited by 120
Omnidirectional DSO: Direct Sparse Odometry With Fisheye Cameras
Matsuki, H., von Stumberg, L., Usenko, V., Stuckler, J., & Cremers, D. (2018) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2018.2855443 · Cited by 98
Chiang, K., Tsai, G.-J., Chu, H., & El‐Sheimy, N. (2020) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2020.2966765 · Cited by 98
State of the Art in Vision-Based Localization Techniques for Autonomous Navigation Systems
Alkendi, Y., Seneviratne, L. D., & Zweiri, Y. (2021) · IEEE Access · doi:10.1109/access.2021.3082778 · Cited by 114
Edge Assisted Mobile Semantic Visual SLAM
Xu, J., Cao, H., Li, D., Huang, K., Qian, C., Shangguan, L., & Yang, Z. (2020) · IEEE INFOCOM 2020 - IEEE Conference on Computer Communications · doi:10.1109/infocom41043.2020.9155438 · Cited by 101
“As another promising alternative, vision-based indoor localization techniques, in particular visual simultaneous localization and mapping (visual SLAM) attracts more attentions in recent years [11], [12].”
BackgroundMethodADVIO: An Authentic Dataset for Visual-Inertial Odometry
Cortés, S., Solin, A., Rahtu, E., & Kannala, J. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01249-6_26 · Cited by 98
PLF-VINS: Real-Time Monocular Visual-Inertial SLAM With Point-Line Fusion and Parallel-Line Fusion
Lee, J., & Park, S.-Y. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3095518 · Cited by 85
Lajoie, P.-Y., Ramtoula, B., Wu, F., & Beltrame, G. (2021) · Field Robotics · doi:10.55417/fr.2022032 · Cited by 105
“The same idea is explored in (Li et al., 2018), in which a popular single-robot SLAM technique (Mur-Artal and Tardós, 2017) is converted into C-SLAM.”
MethodNDT-LOAM: A Real-Time Lidar Odometry and Mapping With Weighted NDT and LFA
Chen, S., Ma, H., Jiang, C., Zhou, B., Xue, W., Xiao, Z., & Li, Q. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3135055 · Cited by 99
Closed-form preintegration methods for graph-based visual–inertial navigation
Eckenhoff, K., Geneva, P., & Huang, G. P. (2019) · The International Journal of Robotics Research · doi:10.1177/0278364919835021 · Cited by 101
“An example of state-of-the-art indirect visual-SLAM methods is ORB-SLAM2 (Mur-Artal and Tard´os, 2017), which performs graph-based optimization of camera poses using information from 3D feature point correspondences.”
MethodImproving monocular visual SLAM in dynamic environments: an optical-flow-based approach
Cheng, J., Sun, Y., & Meng, M. Q. ‐. (2019) · Advanced Robotics · doi:10.1080/01691864.2019.1610060 · Cited by 95
RGBD-Inertial Trajectory Estimation and Mapping for Ground Robots
Shan, Z., Li, R., & Schwertfeger, S. (2019) · Italian National Conference on Sensors · doi:10.3390/s19102251 · Cited by 97
“Visual SLAM with RGBD cameras [13,33] can be applied to robot autonomy and navigation.”
MethodEdge-SLAM: Edge-Assisted Visual Simultaneous Localization and Mapping
Ali, A. J. B., Kouroshli, M., Semenova, S., Hashemifar, Z. S., Ko, S. Y., & Dantu, K. (2022) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3561972 · Cited by 96
“[59] built a collaborative monocular SLAM on top of ORB-SLAM2 [50] for Unmanned Aerial Vehicles ( UAVs ).”
BackgroundMethodInfluentialShu, F., Wang, J., Pagani, A., & Stricker, D. (2023) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48891.2023.10160452 · Cited by 96
“SLAM2 [33] with high usability and extensibility.”
InfluentialSemantic–geometric visual place recognition: a new perspective for reconciling opposing views
Garg, S., Sünderhauf, N., & Milford, M. J. (2019) · The International Journal of Robotics Research · doi:10.1177/0278364919839761 · Cited by 95
“We use the conventional coarse-to-fine matching process, often employed in literature (Cummins and Newman 2011; Mur-Artal and Tardós 2017), for high-precision loop closure detection.”
MethodMixed reality and remote sensing application of unmanned aerial vehicle in fire and smoke detection
Esfahlani, S. S. (2019) · Journal of Industrial Information Integration · doi:10.1016/j.jii.2019.04.006 · Cited by 97
DOT: Dynamic Object Tracking for Visual SLAM
Ballester, I., Fontan, A., Civera, J., Strobl, K. H., & Triebel, R. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561452 · Cited by 94
“A usual approach [5], [1] is modeling dynamic areas as outliers, ignoring them during the pose tracking and map estimation processes.”
BackgroundMethodInfluentialVisual navigation and docking for a planar type AUV docking and charging system
Wang, T., Zhao, Q.-C., & Yang, C.-J. (2021) · Ocean Engineering · doi:10.1016/j.oceaneng.2021.108744 · Cited by 91
VPS-SLAM: Visual Planar Semantic SLAM for Aerial Robotic Systems
Bavle, H., de la Puente, P., How, J., & Campoy, P. (2020) · IEEE Access · doi:10.1109/access.2020.2983121 · Cited by 81
“Most semi-dense SLAM techniques, like [1]–[3], rely on low level characteristic features of the environment such as points, lines and planes.”
BackgroundMethodIntermittent GPS-aided VIO: Online Initialization and Calibration
Lee, W., Eckenhoff, K., Geneva, P., & Huang, G. P. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9197029 · Cited by 92
“A commonly used approach to bound navigation error is a simultaneous localization and mapping (SLAM) that exploits loop-closure constraints to correct the accumulated error [6], [7].”
MethodDSP-SLAM: Object Oriented SLAM with Deep Shape Priors
Wang, J., Rünz, M., & Agapito, L. (2021) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv53792.2021.00143 · Cited by 94
Zhang, Y., Shi, P., & Li, J. (2024) · The Photogrammetric Record · doi:10.1111/phor.12497 · Cited by 103
Canny-VO: Visual Odometry With RGB-D Cameras Based on Geometric 3-D–2-D Edge Alignment
Zhou, Y. B., Li, H., & Kneip, L. (2018) · IEEE Transactions on Robotics · doi:10.1109/tro.2018.2875382 · Cited by 93
Run Your Visual-Inertial Odometry on NVIDIA Jetson: Benchmark Tests on a Micro Aerial Vehicle
Jeon, J., Jung, S., Lee, E. M., Choi, D., & Myung, H. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3075141 · Cited by 96
“We test three mono-VIO (VINS-Mono [16], ROVIO [17], and ALVIO [18]), two stereoVO (ORB-SLAM2 stereo [19] and VINS-Fusion w/o IMU [20]), and four stereo-VIO (VINS-Fusion w/ IMU [20], VINS-Fusion w/ GPU [21], Stereo-MSCKF [22], and Kimera [23]) algorithms and benchmark them on NVIDIA Jetson TX2, Xavier NX, and AGX Xavie…”
SVT-Net: Super Light-Weight Sparse Voxel Transformer for Large Scale Place Recognition
Fan, Z., Song, Z., Liu, H., Lu, Z., He, J., & Du, X. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i1.19934 · Cited by 99
“Large scale place recognition is the spine of a wide range of applications like Simultaneous Localization and Mapping (SLAM) (Mur-Artal and Tardós 2017), Autonomous Driving (Levinson et al. 2011), Robot Navigation (Ravankar et al. 2018), etc. Commonly, the place recognition result can be used for…”
BackgroundInfluentialMAVBench: Micro Aerial Vehicle Benchmarking
Boroujerdian, B., Genc, H., Krishnan, S., Cui, W., Faust, A., & Reddi, V. J. (2018) · IEEE/ACM International Symposium on Microarchitecture (MICRO) · doi:10.1109/micro.2018.00077 · Cited by 92
“These applications’ dataflows are comprised of several state-of-the-art computational kernels, such as object detection [11], [12], occupancy map generation [13], motion planning [14], localization [15], [16], which we integrated together to create complete applications.”
BackgroundMethodLi, L., Li, X., Yang, S., Ding, S., Jolfaei, A., & Zheng, X. (2020) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2020.3011067 · Cited by 92
DI-Fusion: Online Implicit 3D Reconstruction with Deep Priors
Huang, J., Huang, S.-S., Song, H., & Hu, S. (2021) · 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/cvpr46437.2021.00882 · Cited by 93
“The early techniques, such as MonoSLAM [13], LSD-SLAM [15], ORB-SLAM2 [30], and DSO [14], propose to perform camera tracking using sparse features or direct intensity RGB data.”
MethodThe Rosario dataset: Multisensor data for localization and mapping in agricultural environments
Pire, T., Mujica, M., Civera, J., & Kofman, E. (2019) · The International Journal of Robotics Research · doi:10.1177/0278364919841437 · Cited by 91
Xu, L., Feng, C., Kamat, V. R., & Menassa, C. C. (2019) · Automation in Construction · doi:10.1016/j.autcon.2019.04.011 · Cited by 90
Beyond Photometric Loss for Self-Supervised Ego-Motion Estimation
Shen, T., Luo, Z., Zhou, L., Deng, H., Zhang, R., Fang, T., & Quan, L. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8793479 · Cited by 91
“, which is widely used for SfM. The average feature number for each image is around 8000. For weaklysupervised poses, we use the consecutive motion generated by PnP algorithm used in stereo ORB-SLAM2 [31], which is essentially EPnP [23] with RANSAC [9]. We choose the stereo version rather than the monocular one becaus…”
MethodInfluentialDynamic Scene Semantics SLAM Based on Semantic Segmentation
Han, S., & Xi, Z. (2020) · IEEE Access · doi:10.1109/access.2020.2977684 · Cited by 91
“Over the past 30 years, many scholars have carried out research on SLAM and achieved outstanding results, making the SLAM system more mature and able to make a good performance, such as ORB-SLAM2 [1], RGBD-SLAM-V2 [2].”
BackgroundSurvey on Localization Systems and Algorithms for Unmanned Systems
Yuan, S., Wang, H., & Xie, L. (2020) · Unmanned Systems · doi:10.1142/s230138502150014x · Cited by 91
ManhattanSLAM: Robust Planar Tracking and Mapping Leveraging Mixture of Manhattan Frames
Yunus, R., Li, Y., & Tombari, F. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9562030 · Cited by 92
“Additionally, we disable the bundle adjustment and loop closure modules of ORB-SLAM2 and SP-SLAM for a fair comparison.”
BackgroundMethodInfluentialAitken, 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 103
“Popular recent approaches to VO can be divided into feature-based methods, which use feature extraction to obtain image frame correspondences, such as ORB-SLAM [88], [93], and direct methods, which operate directly on pixel intensity, such as large scale direct SLAM (LSD-SLAM) [95] and direct sparse odometry (DSO) [92…”
BackgroundMethodInfluentialHu, X., & Assaad, R. H. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.120897 · Cited by 98
Reinforced Feature Points: Optimizing Feature Detection and Description for a High-Level Task
Bhowmik, A., Gumhold, S., Rother, C., & Brachmann, E. (2020) · 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/cvpr42600.2020.00500 · Cited by 91
“mages has been a long-standing problem in computer vision [19]. Feature detection algorithms enable the creation of vivid 3D models from image collections [21,56,45], building maps for robotic agents [31,32], recognizing places [44,24,35] and precise locations [25,52,41] as well as recognizing objects [26,34,38,1,2].…”
BackgroundHigh-quality indoor scene 3D reconstruction with RGB-D cameras: A brief review
Li, J., Gao, W., Wu, Y., Liu, Y.-D., & Shen, Y.-F. (2022) · Computational Visual Media · doi:10.1007/s41095-021-0250-8 · Cited by 99
“Point features are a popular choice for feature extraction and matching in 3D reconstruction, such as SIFT [34], FPFH [3], ORB [101], and some learned features (e.g., 3DMatch [87] and PixLoc [102]).”
MethodAutomated vision-based construction progress monitoring in built environment through digital twin
Pal, A., Lin, J. J., Hsieh, S., & Golparvar‐Fard, M. (2023) · Developments in the Built Environment · doi:10.1016/j.dibe.2023.100247 · Cited by 119
Kim, P., Coltin, B., & Kim, H. J. (2018) · IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2018.8463207 · Cited by 82
“Note that the proposed method is less sensitive to the existence of enough textures and brightness in the image than the typical feature-based VO methods [11], [10] since the minimum number of points for estimating translation only is smaller than the number of feature points to determine both rotation and translation.”
MethodLarge-scale, real-time visual–inertial localization revisited
Lynen, S., Zeisl, B., Aiger, D., Bosse, M. C., Hesch, J. A., Pollefeys, M., Siegwart, R., & Sattler, T. (2020) · The International Journal of Robotics Research · doi:10.1177/0278364920931151 · Cited by 91
“Localization enables SLAM systems (Williams et al. 2007; Mur-Artal and Tardós 2017) to detect and handle loop closure events.”
BackgroundMethodMoving Object Segmentation and Detection for Robust RGBD-SLAM in Dynamic Environments
Xie, W., Liu, P. X., & Zheng, M. (2020) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2020.3026803 · Cited by 89
“One is ORB-SLAM2 [4], and our system is based on it.”
BackgroundCrowd-SLAM: Visual SLAM Towards Crowded Environments using Object Detection
Soares, J. C. V., Gattass, M., & Meggiolaro, M. A. (2021) · Journal of Intelligent & Robotic Systems · doi:10.1007/s10846-021-01414-1 · Cited by 89
“There are several Visual SLAM systems in the literature, with high precision and efficiency, for instance, ORB-SLAM2 [1], LSD-SLAM [2], RGBDSLAM [3], and RTAB-Map [4].”
BackgroundMethodYan, Y., Zhang, B., Zhou, J. C., Zhang, Y., & Liu, X. (2022) · Agronomy · doi:10.3390/agronomy12081740 · Cited by 90
A review of visual SLAM for robotics: evolution, properties, and future applications
Al-Tawil, B., Hempel, T., Abdelrahman, A. A., & Al-Hamadi, A. K. (2024) · Frontiers in Robotics and AI · doi:10.3389/frobt.2024.1347985 · Cited by 92
Sun, L., Yan, Z., Zaganidis, A., Zhao, C., & Duckett, T. (2018) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2018.2856268 · Cited by 91
Chen, F., Lyu, H., Pang, Z., Zhang, J., Hou, Y., Gu, Y., Yang, H., & Yang, G. (2018) · IEEE Sensors Journal · doi:10.1109/jsen.2018.2877978 · Cited by 88
“the wrist camera includes the background scenes, from which the hand’s velocity and even the position can be estimated by using simultaneous localization and mapping (SLAM) techniques [14].”
MethodLearning Rigidity in Dynamic Scenes with a Moving Camera for 3D Motion Field Estimation
Lv, Z., Kim, K., Troccoli, A. J., Sun, D., Rehg, J. M., & Kautz, J. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01228-1_29 · Cited by 88
Vision-Assisted Autonomous Lower-Limb Exoskeleton Robot
Liu, D.-X., Xu, J. L., Chen, C., Long, X., Tao, D., & Wu, X. (2019) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2019.2932892 · Cited by 88
“In [35], an RGBD camera was used to build the simultaneous localization and mapping (SLAM) system with map reuse, loop closing, and relocalization capabilities.”
MethodLinear RGB-D SLAM for Planar Environments
Kim, P., Coltin, B., & Kim, H. J. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01225-0_21 · Cited by 86
“Many typical visual RGB-D SLAM approaches such as DVO-SLAM [17] and ORB-SLAM2 [23], which are based on the pose graph optimization [19], have shown promising results in the environments with rich texture.”
BackgroundMethodXTDrone: A Customizable Multi-rotor UAVs Simulation Platform
Xiao, K., Tan, S., Wang, G., Xueyan, A., Wang, X., & Wang, X. (2020) · International Conference on Robotics and Automation Sciences (ICRAS) · doi:10.1109/icras49812.2020.9134922 · Cited by 99
“6 shows the trajectories generated by ORB-SLAM2 and VINS-Fusion without IMU, compared with the true flight trajectory.”
MethodInfluentialSniffy Bug: A Fully Autonomous Swarm of Gas-Seeking Nano Quadcopters in Cluttered Environments
Duisterhof, B. P., Li, S., Burgués, J., Reddi, V. J., & de Croon, G. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9636217 · Cited by 87
“State-of-the-art methods use heavy sensors like LiDAR and high-resolution cameras to construct a detailed metric map of the environment, while also estimating the robot’s position for navigation with Simultaneous Localization And Mapping (SLAM, e.g., ORB-SLAM2 [3]).”
Webb, B. H. (2019) · Journal of Experimental Biology · doi:10.1242/jeb.188094 · Cited by 87
“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).”
BackgroundPoint-Plane SLAM Using Supposed Planes for Indoor Environments
Zhang, X., Wang, W., Qi, X., Liao, Z., & Wei, R. (2019) · Sensors · doi:10.3390/s19173795 · Cited by 86
“For implementation, our system augments the RGB-D variant of ORB-SLAM2 [7].”
BackgroundMethodInfluentialJi, P., Li, R., Bhanu, B., & Xu, Y. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01255 · Cited by 89
“Some other methods then resort to generating pseudo ground-truth depth labels with traditional 3D reconstruction meth-ods [23, 22], such as SfM [31] and SLAM [26], or 3D movies [28].”
BackgroundMethodNarazaki, Y., Hoskere, V., Chowdhary, G. V., & Spencer, B. F. (2022) · Automation in Construction · doi:10.1016/j.autcon.2022.104214 · Cited by 85
Edge Robotics: Edge-Computing-Accelerated Multirobot Simultaneous Localization and Mapping
Huang, P., Zeng, L., Chen, X., Huang, L., Zhou, Z., & Yu, S. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3146461 · Cited by 88
ORBSLAM-Atlas: a robust and accurate multi-map system
Elvira, R., Tardós, J. D., & Montiel, J. M. M. (2019) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros40897.2019.8967572 · Cited by 85
“More specifically we build on top of the reference system ORBSLAM [1], [2], [3].”
MethodProgressively Optimized Local Radiance Fields for Robust View Synthesis
Meuleman, A., Liu, Y.-L., Gao, C., Huang, J., Kim, C., Kim, M. H., & Kopf, J. (2023) · 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/cvpr52729.2023.01587 · Cited by 99
“They can either rely directly on the color by maximizing the photoconsistency [46, 51] or on extracted hand-crafted features [25, 26, 34].”
BackgroundUnsupervised Deep Visual-Inertial Odometry with Online Error Correction for RGB-D Imagery
Shamwell, E. J., Lindgren, K., Leung, S., & Nothwang, W. D. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2909895 · Cited by 86
“ORB-SLAM2 [20] is a VSLAM system for monocular, stereo, and RGB-D cameras.”
BackgroundMethodReal-time Scalable Dense Surfel Mapping
Wang, K., Gao, F., & Shen, S. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8794101 · Cited by 83
“Due to the fact that ORB-SLAM2 optimizes the whole pose-graph frequently, our system deforms the map accordingly to maintain global consistency.”
BackgroundInfluential4Seasons: A Cross-Season Dataset for Multi-Weather SLAM in Autonomous Driving
Wenzel, P., Wang, R., Yang, N., Cheng, Q., Khan, Q., von Stumberg, L., Zeller, N., & Cremers, D. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-71278-5_29 · Cited by 84
“During the last decade, research on visual odometry (VO) and simultaneous localization and mapping (SLAM) has made tremendous strides [30,12,29,11] particularly in the context of autonomous driving (AD) [9,44,46,28].”
BackgroundEfficient and Accurate Tightly-Coupled Visual-Lidar SLAM
Chou, C., & Chou, C. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3130089 · Cited by 84
“To facilitate implementation, we use the ORB-SLAM2 [4] source code with minor changes, as in this paper we focus on the integration of visual and lidar SLAM rather than individual single-modality SLAM systems.”
BackgroundMethodAirSLAM: An Efficient and Illumination-Robust Point-Line Visual SLAM System
Xu, K., Hao, Y., Yuan, S., Wang, C., & Xie, L. (2025) · IEEE Transactions on Robotics · doi:10.1109/tro.2025.3539171 · Cited by 89
“Weskipthedetailsof3-Dpoint processinginoursystemastheyaresimilartootherpoint-based VOandSLAMsystems[4],[74].”
Semantic mapping for orchard environments by merging two‐sides reconstructions of tree rows
Dong, W., Roy, P., & Isler, V. (2019) · Journal of Field Robotics · doi:10.1002/rob.21876 · Cited by 83
“, 2011), SfM (Wu, 2013) and SLAM (Mur-Artal and Tardós, 2017) are not reliable enough for tree volume and trunk diameter estimation.”
BackgroundMethodClusterSLAM: A SLAM Backend for Simultaneous Rigid Body Clustering and Motion Estimation
Huang, J., Yang, S. Y., Zhao, Z., Lai, Y., & Hu, S. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00597 · Cited by 82
Ellipse R-CNN: Learning to Infer Elliptical Object From Clustering and Occlusion
Dong, W., Roy, P., Peng, C., & Isler, V. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3050673 · Cited by 83
“While some efforts have been made for 3D fruit localization [25]–[28] by utilizing standard mapping techniques [29], [30], none of them could perform object size estimation because of low-resolution 3D reconstructions.”
BackgroundLidar-Monocular Visual Odometry using Point and Line Features
Huang, S.-S., Ma, Z., Mu, T., Fu, H., & Hu, S. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9196613 · Cited by 83
“We form a point-line bundle adjustment in the back-end between keyframes of a sliding neighbor window, similar to ORB-SLAM2 [7].”
MethodInfluentialSelf-supervised Monocular Depth Estimation for All Day Images using Domain Separation
Liu, L., Song, X., Wang, M., Liu, Y., & Zhang, L. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01250 · Cited by 91
“Self-supervised depth estimation has been applied in a wide range of fields such as augmented reality [3][5], 3D reconstruction [17], SLAM [22][30][31] and scene under-* Figure 1.”
MethodA Comprehensive Review on 3D Object Detection and 6D Pose Estimation With Deep Learning
Hoque, S., Arafat, M. Y., Xu, S., Maiti, A., & Wei, Y.-C. (2021) · IEEE Access · doi:10.1109/access.2021.3114399 · Cited by 91
Simultaneous Localization and Mapping in the Epoch of Semantics: A Survey
Sualeh, M., & Kim, G.-W. (2018) · International Journal of Control Automation and Systems · doi:10.1007/s12555-018-0130-x · Cited by 85
A Comparison of Modern General-Purpose Visual SLAM Approaches
Merzlyakov, A., & Macenski, S. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9636615 · Cited by 84
“While it supports re-localizing, loop closures, and stereo cameras (in addition to monocular cameras), previous experiments have shown LSD-SLAM achieves lower accuracy than the ORB-SLAM family [15].”
A LiDAR SLAM With PCA-Based Feature Extraction and Two-Stage Matching
Guo, S., Zheng, R. Y., Wang, S., & Wu, Y. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3156982 · Cited by 82
Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass
Yang, J., (Sasha), A. S., Liang, K. J., Henaff, M., Tang, H., Cao, A., Chai, J. Y., Meier, F., & Feiszli, M. (2025) · 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/cvpr52734.2025.02042 · Cited by 89
“The latest multi-view geometry pipelines like COLMAP [44] or OrbSLAM2 [30] incorporate nearly 60 years of compounding engineering improvements, but these approaches still catastrophically fail > 40% of the time on static scenes like ETH-3D [52]), which can actually be considered an easy case due to dense image coverag…”
A comparative analysis of tightly-coupled monocular, binocular, and stereo VINS
Paul, M. K., Wu, K., Hesch, J. A., Nerurkar, E. D., & Roumeliotis, S. I. (2017) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2017.7989022 · Cited by 82
“7 compares the performance of the proposed stereo system (with the FREAK-matching frontend) against two state-of-the-art stereo SLAM systems: i) OKVIS [13] and ii) ORB-SLAM2 [14].”
MethodPyRobot: An Open-source Robotics Framework for Research and BenchmarkingPreprint
Murali, A., Chen, T., Alwala, K. V., Gandhi, D., Pinto, L., Gupta, S., & Gupta, A. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.08236 · Cited by 116
“We deployed ORB-SLAM2 [51], Figure 4: Qualitative comparisons for trajectory tacking for LoCoBot and LoCoBot-Lite.”
MethodZhang, H., Wang, G., Lei, Z., & Hwang, J.-N. (2019) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3343031.3350933 · Cited by 86
“ns; (2) 3D-localized of detected target objects and get the distance from camera to objects; To achieve the first goal, simultaneous localization and mapping (SLAM) technology is introduced. ORB-SLAM [23] is a symbolic framework for computing camera trajectory in real-time by extracting and tracking feature points acr…”
BackgroundDM-SLAM: A Feature-Based SLAM System for Rigid Dynamic Scenes
Cheng, J., Wang, Z., Zhou, H., Li, L., & Yao, J. (2020) · ISPRS International Journal of Geo-Information · doi:10.3390/ijgi9040202 · Cited by 80
“In this paper, a novel visual SLAM method for rigid dynamic scenes (DM-SLAM) is proposed based on the ORB-SLAM2 system [3].”
MethodLearning Monocular Visual Odometry via Self-Supervised Long-Term Modeling
Zou, Y., Ji, P., Tran, Q.-H., Huang, J., & Chandraker, M. K. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58568-6_42 · Cited by 83
Xu, H., Liu, P., Chen, X., & Shen, S. (2024) · IEEE Transactions on Robotics · doi:10.1109/tro.2024.3422003 · Cited by 92
“SuperPoint is preferred for its robustness over ORB features [50].”
BackgroundMethodPLGSLAM: Progressive Neural Scene Represenation with Local to Global Bundle Adjustment
Deng, T., Shen, G., Qin, T., Wang, J., Zhao, W., Wang, J., Wang, D., & Chen, W. (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/cvpr52733.2024.01859 · Cited by 94
“Many traditional methods have been introduced in the past years, such as ORB-SLAM [19, 20], VINS [24], and so on.”
Omnidirectional 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 79
3D Mapping and 6D Pose Computation for Real Time Augmented Reality on Cylindrical Objects
Tang, F.-L., Wu, Y., Hou, X., & Ling, H. (2020) · IEEE transactions on circuits and systems for video technology (Print) · doi:10.1109/tcsvt.2019.2950449 · Cited by 90
“Also, we utilize ORBSLAM2 [32] and COLMAP [33] to reconstruct the 3D models, which are shown in (b) and (c) of Figure 8.”
MethodAdvances in Inference and Representation for Simultaneous Localization and Mapping
Rosen, D. M., Doherty, K., Espinoza, A. T., & Leonard, J. J. (2021) · Annual Review of Control Robotics and Autonomous Systems · doi:10.1146/annurev-control-072720-082553 · Cited by 91
“ORB-SLAM: a versatile and accurate monocular SLAM system.”
BackgroundMethodInfluentialWang, C., Xu, A., Kuang, J., Sui, X., Hao, Y., & Niu, X. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3082579 · Cited by 79
Metadata
- Authors
- Mur-Artal, R., & Tardos, J. D.
- Affiliations
- Instituto de Investigación en Ingeniería de Aragón (ES)
- Venue
- IEEE Transactions on Robotics
- Published
- 2017-10
- Volume
- 33
- Issue
- 5
- Pages
- 1255-1262
- ISSN
- 1552-3098 (ISSN-L)
- 1941-0468
- OpenAlex
- W2535547924
- Cited by
- 6,898
- References
- 21
- Citations by year
- 2026: 521 · 2025: 851 · 2024: 944 · 2023: 962 · 2022: 910 · 2021: 870 · 2020: 754 · 2019: 636 · 2018: 360 · 2017: 90
- FWCI
- 463.87
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Transactions on Robotics (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Spanish government (DPI2015-67275)
- Aragón regional government (DGA T04-FSE)
- Ministerio de Educación Scholarship (FPU13/04175)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Earth and Planetary Sciences › Geology › 3D Surveying and Cultural Heritage
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
- Keywords
- ORB-SLAM2, SLAM, monocular SLAM, stereo SLAM, RGB-D SLAM, loop closure, relocalization, bundle adjustment, visual odometry, trajectory estimation
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
