ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual–Inertial, and Multimap SLAM
By Carlos Campos, Richard Elvira, Juan J. Gómez Rodríguez, José M. M. Montiel, Juan D. Tardós. IEEE Transactions on robotics, 2020
4,364 citations recorded by the source corpus.
This article presents ORB-SLAM3, the first system able to perform visual, visual-inertial and multimap SLAM with monocular, stereo and RGB-D cameras, using pin-hole and fisheye lens models. The first main novelty is a tightly integrated visual-inertial SLAM system that fully relies on maximum a posteriori (MAP) estimation, even during IMU initialization, resulting in real-time robust operation in small and large, indoor and outdoor environments, being two to ten times more accurate than previous approaches. The second main novelty is a multiple map system relying on a new place recognition method with improved recall that lets ORB-SLAM3 survive to long periods of poor visual information: when it gets lost, it starts a new map that will be seamlessly merged with previous maps when revisiting them. Compared with visual odometry systems that only use information from the last few seconds, ORB-SLAM3 is the first system able to reuse in all the algorithm stages all previous information from high parallax co-visible keyframes, even if they are widely separated in time or come from previous mapping sessions, boosting accuracy. Our experiments show that, in all sensor configurations, ORB-SLAM3 is as robust as the best systems available in the literature and significantly more accurate. Notably, our stereo-inertial SLAM achieves an average accuracy of 3.5 cm in the EuRoC drone and 9 mm under quick hand-held motions in the room of TUM-VI dataset, representative of AR/VR scenarios. For the benefit of the community we make public the source code.
Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Geology
View Paper: https://www.semanticscholar.org/paper/77d7e439b3368875199a1327515a3ba212f0a359
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Aerospace Engineering
- Luca Carlone
- Antoni Rosinol
- Yohann Cabon
- Vincent Leroy
- Jérôme Revaud
- Yun Chang
- Zachary Teed
- Jia Deng
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Andrea Vedaldi
- Dacheng Tao
- Jitendra Malik
- Xiaogang Wang
- Cordelia Schmid
- Andrew Zisserman
- Trevor Darrell
In Geology
Citations in the corpus
- Adaptive VIO: Deep Visual-Inertial Odometry with Online Continual Learning (2024, methodology): “The IMU timestamps and bias configuration follow the settings in [6, 24].”
- PGSR: Planar-Based Gaussian Splatting for Efficient and High-Fidelity Surface Reconstruction (2024, positive): “This paper focuses exclusively on surface reconstruction under given poses, which can be readily computed using SLAM [25], [26], [27] or SFM [28], [29], [30] methods.”
- DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras (2021, methodology): “Stereo OV 2 SLAM [17] 0.510 0.182 VOLDOR [28] + COLMAP [41] 0.440 0.177 SuperGlue [39] + SuperPoint [13] + COLMAP [41] In the monocular setting, we achieve an average ATE of 2.2cm, reducing error by 82% among methods with zero failures, and by 43% over ORB-SLAM3 when only comparing sequences where O”
- DUSt3R: Geometric 3D Vision Made Easy (2023, neutral): “On top of that, key steps in this pipeline are brittle and prone to break in many cases [59].”
- Kimera: From SLAM to spatial perception with 3D dynamic scene graphs (2021, methodology): “While most open-source SLAM algorithms, such as ORBSLAM3 (Campos et al., 2021), VINS-Mono (Qin et al.”
- A Comprehensive Survey of Visual SLAM Algorithms (2022, methodology): “In a general analysis, the addition of an IMU to visual-based SLAM algorithms has the primary purpose of increasing the system’s robustness, which was already demonstrated to be true [2,22,70].”
- SplaTAM: Splat, Track & Map 3D Gaussians for Dense RGB-D SLAM (2023, methodology): “For the past three decades, SLAM research has extensively centered around the question of map representation – resulting in a variety of sparse [2, 3, 7, 23], dense [4, 6, 8, 13, 15, 25, 26, 34, 42, 43], and neural scene representations [21, 29, 30, 37, 46, 55].”
- NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Video (2021, background): “While camera motion can be tracked accurately with state-of-the-art visual-inertial SLAM systems [3, 35, 1], real-time image-based dense reconstruction remains to be a challenging problem due to low reconstruction quality and high computation demands.”
- Grounding Image Matching in 3D with MASt3R (2024, background): “…correspondences between pixels across different images of the same scene, denoted as image matching , constitutes a core component of all 3D vision applications, spanning mapping [14,61], localization [41,72], navigation [15], photogrammetry [34, 64] and autonomous robotics in general [63,87].”