ORB-SLAM3

ORB-SLAM3 is a real-time SLAM library for building maps and tracking camera position using visual, visual-inertial, and multi-map methods.

Made by SLAMLab, Universidad de Zaragoza, Spain (checked on github.com/UZ-SLAMLab (opens in a new tab)).

Licenses

Code: GPL-3.0-only (opens in a new tab)
Read by Robotheca from LICENSE (opens in a new tab) (GPL-3.0-only) at commit 4452a3c (opens in a new tab), on 16 Oct 2026.
Commercial use: Yes

Copyleft: Modified source must be published.

Performs visual, visual-inertial and multi-map SLAM in real time.
Works with monocular, stereo and RGB-D cameras using pin-hole or fisheye lens models.
Includes example runs on the EuRoC and TUM-VI datasets, with and without an IMU.
Provides demos for Intel Realsense T265 and D435i cameras and a guide for calibrating your own camera.
Offers optional ROS nodes for monocular, monocular-inertial, stereo, stereo-inertial and RGB-D inputs.
Can measure running time per stage when the REGISTER_TIMES flag is enabled in Config.h.

Written by Robotheca from the project's README (opens in a new tab) at commit 4452a3c.

Difficulty: Advanced
Hardware it needs, as the project states it: A powerful computer such as an i7 is suggested to ensure real-time performance. No GPU or memory figures are given.
How to start: The project's README (opens in a new tab)

Checked on GitHub on 16 Oct 2026.

No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.

Robotheca lists software as a reference: nothing is hosted or run here, and every fact links to where it was read.