Skip to main content

ORB-SLAM3 Documentation

Learn how to integrate and use ORB-SLAM3, the first real-time SLAM library supporting Visual, Visual-Inertial and Multi-Map SLAM with monocular, stereo and RGB-D cameras.

Get StartedAPI Reference

Quick Start

Get ORB-SLAM3 up and running in your project

1

Install Dependencies

ORB-SLAM3 requires C++11, OpenCV (3.0+), Eigen3 (3.1.0+), and Pangolin for visualization.
sudo apt install libopencv-dev libeigen3-dev
2

Clone and Build

Clone the repository and build the library along with third-party dependencies.
git clone https://github.com/UZ-SLAMLab/ORB_SLAM3.git ORB_SLAM3
cd ORB_SLAM3
chmod +x build.sh
./build.sh
This creates libORB_SLAM3.so in the lib folder and example executables in the Examples folder.
3

Run Your First Example

Test the system with a sample dataset. Download an EuRoC sequence and run:
./Examples/Stereo/stereo_euroc Vocabulary/ORBvoc.txt Examples/Stereo/EuRoC.yaml /path/to/sequence/mav0/cam0/data /path/to/sequence/mav0/cam1/data Examples/Stereo/EuRoC_TimeStamps/SEQUENCE.txt

Explore by Topic

Deep dive into ORB-SLAM3’s capabilities

Core Concepts

Understand SLAM fundamentals and ORB-SLAM3’s architecture

Sensor Modes

Learn about monocular, stereo, RGB-D, and inertial configurations

Camera Models

Work with pin-hole and fisheye lens models

Configuration

Configure YAML settings for your camera and sensor setup

Calibration

Calibrate your camera and IMU sensors

ROS Integration

Integrate ORB-SLAM3 with ROS for real-time robotics applications

Examples & Datasets

Run ORB-SLAM3 with standard benchmarks and live cameras

EuRoC Dataset

Visual-inertial sequences with ground truth trajectories

TUM-VI Dataset

Fisheye camera sequences for challenging environments

KITTI Dataset

Autonomous driving sequences with stereo cameras

RealSense Cameras

Live SLAM with Intel RealSense D435i and T265

API Reference

Complete reference for ORB-SLAM3 classes and methods

System

Main interface for initializing and running ORB-SLAM3 with different sensor configurations

View documentation

Tracking

Camera pose estimation and feature tracking for real-time localization

View documentation

Atlas

Multi-map management system for seamless map fusion and switching

View documentation

IMU Types

Inertial measurement data structures and integration methods

View documentation

Resources

Additional materials and community support

Research Paper

Read the IEEE Transactions on Robotics paper describing ORB-SLAM3

GitHub Repository

Browse the source code, report issues, and contribute

Calibration Tutorial

Detailed PDF guide for camera and IMU calibration

Performance Analysis

Tools and methods for evaluating SLAM accuracy

Ready to Get Started?

Build accurate real-time SLAM applications with ORB-SLAM3’s powerful visual and inertial tracking capabilities.

Install ORB-SLAM3View Examples