ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Cameras
arXiv:1610.06475 · doi:10.1109/TRO.2017.2705103
Abstract
We present ORB-SLAM2 a complete 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 CPUs 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 to 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.
arXiv commentAccepted for publication in IEEE Transactions on Robotics
References in corpus (2)
Cited by in corpus (228)
- ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM
- DynaSLAM: Tracking, Mapping and Inpainting in Dynamic Scenes
- DS-SLAM: A Semantic Visual SLAM towards Dynamic Environments
- RTAB-Map as an Open-Source Lidar and Visual SLAM Library for Large-Scale and Long-Term Online Operation
- Visual-Inertial Monocular SLAM with Map Reuse
- Seeing biodiversity: perspectives in machine learning for wildlife conservation
- Grape detection, segmentation and tracking using deep neural networks and three-dimensional association
- Intensity Scan Context: Coding Intensity and Geometry Relations for Loop Closure Detection
- OpenVSLAM: A Versatile Visual SLAM Framework
- maplab: An Open Framework for Research in Visual-inertial Mapping and Localization
- Event-based Stereo Visual Odometry
- Vox-Fusion: Dense Tracking and Mapping with Voxel-based Neural Implicit Representation
- Metric3Dv2: A Versatile Monocular Geometric Foundation Model for Zero-shot Metric Depth and Surface Normal Estimation
- Unsupervised Scale-consistent Depth Learning from Video
- Blind MultiChannel Identification and Equalization for Dereverberation and Noise Reduction based on Convolutive Transfer Function
- Joint Alignment of Multiple Point Sets with Batch and Incremental Expectation-Maximization
- Omni-swarm: A Decentralized Omnidirectional Visual-Inertial-UWB State Estimation System for Aerial Swarms
- Perception and Navigation in Autonomous Systems in the Era of Learning: A Survey
- Reference Pose Generation for Long-term Visual Localization via Learned Features and View Synthesis
- Track to Reconstruct and Reconstruct to Track
- GANVO: Unsupervised Deep Monocular Visual Odometry and Depth Estimation with Generative Adversarial Networks
- Virtual-Blind-Road Following Based Wearable Navigation Device for Blind People
- Event-aided Direct Sparse Odometry
- Visual SLAM: What are the Current Trends and What to Expect?
- EAO-SLAM: Monocular Semi-Dense Object SLAM Based on Ensemble Data Association
- BoW3D: Bag of Words for Real-Time Loop Closing in 3D LiDAR SLAM
- Towards Real-time Semantic RGB-D SLAM in Dynamic Environments
- RGB-D Inertial Odometry for a Resource-Restricted Robot in Dynamic Environments
- A Survey on Approximate Edge AI for Energy Efficient Autonomous Driving Services
- A Region-based Gauss-Newton Approach to Real-Time Monocular Multiple Object Tracking
- Omnidirectional DSO: Direct Sparse Odometry with Fisheye Cameras
- Canny-VO: Visual Odometry with RGB-D Cameras based on Geometric 3D-2D Edge Alignment
- Local Feature Matching Using Deep Learning: A Survey
- Ellipse R-CNN: Learning to Infer Elliptical Object from Clustering and Occlusion
- Ultra-Wideband Aided Fast Localization and Mapping System
- Semantic Mapping for Orchard Environments by Merging Two-Sides Reconstructions of Tree Rows
- Closed-form Preintegration Methods for Graph-based Visual-Inertial Navigation
- AirSLAM: An Efficient and Illumination-Robust Point-Line Visual SLAM System
- LiDAR-Camera Calibration under Arbitrary Configurations: Observability and Methods
- Eye in the Sky: Drone-Based Object Tracking and 3D Localization
- maplab 2.0 -- A Modular and Multi-Modal Mapping Framework
- Masked GANs for Unsupervised Depth and Pose Prediction with Scale Consistency
- SGANVO: Unsupervised Deep Visual Odometry and Depth Estimation with Stacked Generative Adversarial Networks
- From SLAM to Situational Awareness: Challenges and Survey
- Active SLAM: A Review On Last Decade
- Unsupervised Learning of Depth, Optical Flow and Pose with Occlusion from 3D Geometry
- SRT3D: A Sparse Region-Based 3D Object Tracking Approach for the Real World
- Walking Posture Adaptation for Legged Robot Navigation in Confined Spaces
- Autonomous Hybrid Ground/Aerial Mobility in Unknown Environments
- ProSLAM: Graph SLAM from a Programmer's Perspective
- Loosely-Coupled Semi-Direct Monocular SLAM
- On Deep Learning Techniques to Boost Monocular Depth Estimation for Autonomous Navigation
- AirVO: An Illumination-Robust Point-Line Visual Odometry
- DenseLiDAR: A Real-Time Pseudo Dense Depth Guided Depth Completion Network
- Edge-Assisted Multi-Robot Visual-Inertial SLAM with Efficient Communication
- Gaussian Splatting: 3D Reconstruction and Novel View Synthesis, a Review
- TwistSLAM: Constrained SLAM in Dynamic Environment
- Experimental Comparison of Visual-Aided Odometry Methods for Rail Vehicles
- Deep Direct Visual Odometry
- DASGIL: Domain Adaptation for Semantic and Geometric-aware Image-based Localization
- Delta Descriptors: Change-Based Place Representation for Robust Visual Localization
- Stereo Dense Scene Reconstruction and Accurate Localization for Learning-Based Navigation of Laparoscope in Minimally Invasive Surgery
- Fast ORB-SLAM without Keypoint Descriptors
- HBST: A Hamming Distance embedding Binary Search Tree for Visual Place Recognition
- An Object SLAM Framework for Association, Mapping, and High-Level Tasks
- Navigating the Landscape for Real-time Localisation and Mapping for Robotics and Virtual and Augmented Reality
- Globally Consistent 3D LiDAR Mapping with GPU-accelerated GICP Matching Cost Factors
- Redesigning SLAM for Arbitrary Multi-Camera Systems
- HybVIO: Pushing the Limits of Real-time Visual-inertial Odometry
- Energy-based Legged Robots Terrain Traversability Modeling via Deep Inverse Reinforcement Learning
- JigsawNet: Shredded Image Reassembly using Convolutional Neural Network and Loop-based Composition
- Empty Cities: a Dynamic-Object-Invariant Space for Visual SLAM
- Augmenting Visual Place Recognition with Structural Cues
- Dronument: System for Reliable Deployment of Micro Aerial Vehicles in Dark Areas of Large Historical Monuments
- An Effective Multi-Cue Positioning System for Agricultural Robotics
- Hybrid Visual SLAM for Underwater Vehicle Manipulator Systems
- Feature-based visual odometry prior for real-time semi-dense stereo SLAM
- Balancing the Budget: Feature Selection and Tracking for Multi-Camera Visual-Inertial Odometry
- Direct Sparse Odometry with Rolling Shutter
- Experimental Evaluation of Visual-Inertial Odometry Systems for Arable Farming
- Robust GNSS Denied Localization for UAV Using Particle Filter and Visual Odometry
- Panoramic annular SLAM with loop closure and global optimization
- Rolling-Shutter Modelling for Direct Visual-Inertial Odometry
- Co-Planar Parametrization for Stereo-SLAM and Visual-Inertial Odometry
- Domain-invariant Similarity Activation Map Contrastive Learning for Retrieval-based Long-term Visual Localization
- Panoramic Annular Localizer: Tackling the Variation Challenges of Outdoor Localization Using Panoramic Annular Images and Active Deep Descriptors
- Self-Supervised Scale Recovery for Monocular Depth and Egomotion Estimation
- NeSLAM: Neural Implicit Mapping and Self-Supervised Feature Tracking With Depth Completion and Denoising
- GMMLoc: Structure Consistent Visual Localization with Gaussian Mixture Models
- VIO-DualProNet: Visual-Inertial Odometry with Learning Based Process Noise Covariance
- Visual Semantic SLAM with Landmarks for Large-Scale Outdoor Environment
- A Unified Framework for Mutual Improvement of SLAM and Semantic Segmentation
- Network Uncertainty Informed Semantic Feature Selection for Visual SLAM
- Active View Planning for Visual SLAM in Outdoor Environments Based on Continuous Information Modeling
- HRBF-Fusion: Accurate 3D reconstruction from RGB-D data using on-the-fly implicits
- iSLAM: Imperative SLAM
- Radar-on-Lidar: metric radar localization on prior lidar maps
- Retrieval-based Localization Based on Domain-invariant Feature Learning under Changing Environments
- ViPR: Visual-Odometry-aided Pose Regression for 6DoF Camera Localization
- Minimization of GNSS-Denied Inertial Navigation Errors for Fixed Wing Autonomous Unmanned Air Vehicles
- UAV Autonomous Localization using Macro-Features Matching with a CAD Model
- Transformer-Based Model for Monocular Visual Odometry: A Video Understanding Approach
- RGB-D SLAM in Indoor Planar Environments with Multiple Large Dynamic Objects
- TransFusionOdom: Interpretable Transformer-based LiDAR-Inertial Fusion Odometry Estimation
- Probabilistic Spatial Distribution Prior Based Attentional Keypoints Matching Network
- Sparse Pose Graph Optimization in Cycle Space
- Privacy-Protection Drone Patrol System based on Face Anonymization
- Digital Twin-based 3D Map Management for Edge-assisted Device Pose Tracking in Mobile AR
- Object SLAM-Based Active Mapping and Robotic Grasping
- Deep Online Correction for Monocular Visual Odometry
- RO-MAP: Real-Time Multi-Object Mapping with Neural Radiance Fields
- Visual-Inertial RGB-D SLAM with Encoders for a Differential Wheeled Robot
- GNSS-stereo-inertial SLAM for arable farming
- Augmenting Visual SLAM with Wi-Fi Sensing For Indoor Applications
- SLAM for Visually Impaired People: a Survey
- Stereo Camera Visual SLAM with Hierarchical Masking and Motion-state Classification at Outdoor Construction Sites Containing Large Dynamic Objects
- GOReloc: Graph-based Object-Level Relocalization for Visual SLAM
- On Incremental Structure-from-Motion using Lines
- UNav: An Infrastructure-Independent Vision-Based Navigation System for People with Blindness and Low vision
- Benchmarking Neural Radiance Fields for Autonomous Robots: An Overview
- Online Visual Place Recognition via Saliency Re-identification
- Continual Adaptation of Semantic Segmentation using Complementary 2D-3D Data Representations
- Multitask Learning for Scalable and Dense Multilayer Bayesian Map Inference
- OysterSim: Underwater Simulation for Enhancing Oyster Reef Monitoring
- Indoor simultaneous localization and mapping based on fringe projection profilometry
- Efficient 2D Graph SLAM for Sparse Sensing
- Knowledge Distillation for Feature Extraction in Underwater VSLAM
- Multiple Hypothesis Semantic Mapping for Robust Data Association
- Non-cooperative Stochastic Target Encirclement by Anti-synchronization Control via Range-only Measurement
- FRAME: Fast and Robust Autonomous 3D point cloud Map-merging for Egocentric multi-robot exploration
- Embodied AI with Foundation Models for Mobile Service Robots: A Systematic Review
- Radarize: Enhancing Radar SLAM with Generalizable Doppler-Based Odometry
- Integration of the 3D Environment for UAV Onboard Visual Object Tracking
- Scale-Adaptive Neural Dense Features: Learning via Hierarchical Context Aggregation
- BAMF-SLAM: Bundle Adjusted Multi-Fisheye Visual-Inertial SLAM Using Recurrent Field Transforms
- MM-LINS: a Multi-Map LiDAR-Inertial System for Over-Degenerate Environments
- Unsupervised Visual Odometry and Action Integration for PointGoal Navigation in Indoor Environment
- Deployment of Aerial Robots after a major fire of an industrial hall with hazardous substances, a report
- MG-SLAM: Structure Gaussian Splatting SLAM with Manhattan World Hypothesis
- GS3LAM: Gaussian Semantic Splatting SLAM
- Combining 3D Model Contour Energy and Keypoints for Object Tracking
- Metric Monocular Localization Using Signed Distance Fields
- SymbioLCD: Ensemble-Based Loop Closure Detection using CNN-Extracted Objects and Visual Bag-of-Words
- FRAME: A Modular Framework for Autonomous Map Merging: Advancements in the Field
- The NAO Backpack: An Open-hardware Add-on for Fast Software Development with the NAO Robot
- Monocular visual simultaneous localization and mapping: (r)evolution from geometry to deep learning-based pipelines
- RGB-Only Gaussian Splatting SLAM for Unbounded Outdoor Scenes
- Tightly Coupled Learning Strategy for Weakly Supervised Hierarchical Place Recognition
- Stochastic triangular mesh mapping: A terrain mapping technique for autonomous mobile robots
- Autonomous Robot Navigation with Rich Information Mapping in Nuclear Storage Environments
- RGB-D Odometry and SLAM
- OpenMPR: Recognize Places Using Multimodal Data for People with Visual Impairments
- Evaluating the Impact of Semantic Segmentation and Pose Estimation on Dense Semantic SLAM
- Det-SLAM: A semantic visual SLAM for highly dynamic scenes using Detectron2
- An Addendum to NeBula: Towards Extending TEAM CoSTAR's Solution to Larger Scale Environments
- Tac2Structure: Object Surface Reconstruction Only through Multi Times Touch
- DynaMoN: Motion-Aware Fast and Robust Camera Localization for Dynamic Neural Radiance Fields
- Segment Anything Model is a Good Teacher for Local Feature Learning
- Fully Convolutional Geometric Features for Category-level Object Alignment
- NGD-SLAM: Towards Real-Time Dynamic SLAM without GPU
- ORBBuf: A Robust Buffering Method for Remote Visual SLAM
- LitAR: Visually Coherent Lighting for Mobile Augmented Reality
- Localization and Mapping using Instance-specific Mesh Models
- AirCode: A Robust Object Encoding Method
- Real-time Image Enhancement for Vision-based Autonomous Underwater Vehicle Navigation in Murky Waters
- Marker-based Visual SLAM leveraging Hierarchical Representations
- AG2U -- Autonomous Grading Under Uncertainties
- Map Point Selection for Visual SLAM
- MORPH-DSLAM: Model Order Reduction for PHysics-based Deformable SLAM
- Multi-modal Visual Place Recognition in Dynamics-Invariant Perception Space
- Learning to Segment Dynamic Objects using SLAM Outliers
- Fast Color-guided Depth Denoising for RGB-D Images by Graph Filtering
- GSORB-SLAM: Gaussian Splatting SLAM benefits from ORB features and Transmittance information
- LE-HGR: A Lightweight and Efficient RGB-based Online Gesture Recognition Network for Embedded AR Devices
- GNSS-Denied Semi Direct Visual Navigation for Autonomous UAVs Aided by PI-Inspired Inertial Priors
- DenseSplat: Densifying Gaussian Splatting SLAM with Neural Radiance Prior
- A Fast and Light-weight Non-Iterative Visual Odometry with RGB-D Cameras
- Semihierarchical Reconstruction and Weak-area Revisiting for Robotic Visual Seafloor Mapping
- LiveNVS: Neural View Synthesis on Live RGB-D Streams
- SmartDepthSync: Open Source Synchronized Video Recording System of Smartphone RGB and Depth Camera Range Image Frames with Sub-millisecond Precision
- Long Distance GNSS-Denied Visual Inertial Navigation for Autonomous Fixed Wing Unmanned Air Vehicles: SO(3) Manifold Filter based on Virtual Vision Sensor
- Using Detection, Tracking and Prediction in Visual SLAM to Achieve Real-time Semantic Mapping of Dynamic Scenarios
- A Model for Multi-View Residual Covariances based on Perspective Deformation
- PHYSFRAME: Type Checking Physical Frames of Reference for Robotic Systems
- A comparative evaluation of learned feature descriptors on hybrid monocular visual SLAM methods
- Towards Domain-agnostic Depth Completion
- SplatMAP: Online Dense Monocular SLAM with 3D Gaussian Splatting
- CLIP-Clique: Graph-based Correspondence Matching Augmented by Vision Language Models for Object-based Global Localization
- YoloTag: Vision-based Robust UAV Navigation with Fiducial Markers
- HIPer: A Human-Inspired Scene Perception Model for Multifunctional Mobile Robots
- UAV-assisted Visual SLAM Generating Reconstructed 3D Scene Graphs in GPS-denied Environments
- Line-Circle-Square (LCS): A Multilayered Geometric Filter for Edge-Based Detection
- ATG-PVD: Ticketing Parking Violations on A Drone
- CNN-based local features for navigation near an asteroid
- Design, Implementation and Evaluation of an External Pose-Tracking System for Underwater Cameras
- An Expert Ensemble for Detecting Anomalous Scenes, Interactions, and Behaviors in Autonomous Driving
- Keyframe-based Dense Mapping with the Graph of View-Dependent Local Maps
- CurriculumLoc: Enhancing Cross-Domain Geolocalization through Multi-Stage Refinement
- Fast Loop Closure Detection via Binary Content
- Sparse 3D Point-cloud Map Upsampling and Noise Removal as a vSLAM Post-processing Step: Experimental Evaluation
- Beyond ANN: Exploiting Structural Knowledge for Efficient Place Recognition
- LEVIO: Lightweight Embedded Visual Inertial Odometry for Resource-Constrained Devices
- Enhancing In-vehicle Multiple Object Tracking Systems with Embeddable Ising Machines
- Landmark Stereo Dataset for Landmark Recognition and Moving Node Localization in a Non-GPS Battlefield Environment
- Indoor Localization for an Autonomous Model Car: A Marker-Based Multi-Sensor Fusion Framework
- WildFusion: Multimodal Implicit 3D Reconstructions in the Wild
- Visual Place Representation and Recognition from Depth Images
- Understanding why SLAM algorithms fail in modern indoor environments
- GLC-SLAM: Gaussian Splatting SLAM with Efficient Loop Closure
- Virtual Testbed for Monocular Visual Navigation of Small Unmanned Aircraft Systems
- 3D Freehand Ultrasound using Visual Inertial and Deep Inertial Odometry for Measuring Patellar Tracking
- Are Minimal Radial Distortion Solvers Necessary for Relative Pose Estimation?
- A Novel ViDAR Device With Visual Inertial Encoder Odometry and Reinforcement Learning-Based Active SLAM Method
- SciceVPR: Stable Cross-Image Correlation Enhanced Model for Visual Place Recognition
- PGD-VIO: An Accurate Plane-Aided Visual-Inertial Odometry with Graph-Based Drift Suppression
- RSV-SLAM: Toward Real-Time Semantic Visual SLAM in Indoor Dynamic Environments
- Reproducible Evaluation of Camera Auto-Exposure Methods in the Field: Platform, Benchmark and Lessons Learned
- Embracing Dynamics: Dynamics-aware 4D Gaussian Splatting SLAM
- Splat-based 3D Scene Reconstruction with Extreme Motion-blur
- UDGS-SLAM : UniDepth Assisted Gaussian Splatting for Monocular SLAM
- Visual SLAM-based Localization and Navigation for Service Robots: The Pepper Case
- MSG-Loc: Multi-Label Likelihood-based Semantic Graph Matching for Object-Level Global Localization
- Bias-Eliminated PnP for Stereo Visual Odometry: Provably Consistent and Large-Scale Localization
- iMED: A Multi-Endoscope Dataset for Surgical 3D Perception
- Knowledge Distillation for Underwater Feature Extraction and Matching via GAN-synthesized Images
- Motion Consistency Loss for Monocular Visual Odometry with Attention-Based Deep Learning
- FastTrack: GPU-Accelerated Tracking for Visual SLAM
- ParkingTwin: Training-Free Streaming 3D Reconstruction for Parking-Lot Digital Twins