ai-agent

ORB SLAM2

Name: ORB SLAM2
Author: raulmur

Real-Time SLAM for Monocular, Stereo and RGB-D Cameras, with Loop Detection and Relocalization Capabilities

10,197 stars4,751 forksC++Updated 5/15/2024100% free · open source

What it does

ORB_SLAM2 is a real-time SLAM (Simultaneous Localization and Mapping) system for monocular, stereo, and RGB-D cameras that can detect loops and relocalize, enabling robust and accurate tracking and mapping in various environments.

Install / run

git clone https://github.com/raulmur/ORB_SLAM2.git

When to use it

•When you need to track the movement of a camera in a dynamic environment and build a map of the scene.
•When you want to enable autonomous navigation for robots or drones using visual information.
•When you need to perform 3D reconstruction or augmented reality applications that require accurate camera pose estimation.

Quick start

1Modify the configuration file `config.yaml` to specify the type of camera and other parameters.
2Compile the ORB_SLAM2 library using the command `mkdir build && cd build && cmake .. && make`.
3Run the monocular example using the command `./Examples/Monocular/mono_tum Vocabulary/ORBvoc.txt Examples/Monocular/TUM1.yaml`.
4Use the `mono_tum` executable to process a sequence of images and generate a trajectory and map.
5Visualize the results using a tool like `octomap` or `pcl` to see the reconstructed 3D map.

Ready-to-paste prompt

./Examples/Monocular/mono_tum Vocabulary/ORBvoc.txt Examples/Monocular/TUM1.yaml

Heads up: Make sure you have a compatible C++ compiler (e.g. GCC 5 or later) and the necessary dependencies, including OpenCV and Eigen, installed on your system before compiling ORB_SLAM2.

Saves to your device

How ORB SLAM2 works

Codeflow

Free to inspect

ORB SLAM2 is mostly documents (39 doc files, 53 small scripts) — something you read, not something you run. There is nothing to install. Last commit 28 months ago, Other license.

Size

Mostly documents

39 documents · 53 small scripts · days to read — use, don't read

Setup

Nothing to install

A guide / curated list. Just read it and follow the links.

Runs on

Nowhere — you read it

A reading resource, not a program.

C++ 96%CMake 3%C 1%

Where to start reading