Skip to main content

Luxonis - OAK Series

The OAK cameras by Luxonis are small cameras for robotic vision with an on-device Neural Network inferencing and Computer Vision capabilities. Luxonis is a spatial AI and CV platform that focuses on embedded machine learning and computer vision technology.

  • OAK-1: a compact camera module designed for versatile use in various AI projects. Thanks to the processor, it allows you to detect objects in real time.
  • OAK-D Lite is a more advanced version of OAK-1. It comes with additional features such as an RGB-D depth sensor and stereo cameras.

The advantages of both cameras are their sizes and a good price-to-performance ratio.


Getting started

To run the device, you must meet certain requirements listed below:



  • Connectivity: USB 3.0


In this demo, we'll walk you through using the OAK cameras with ROS 2 via a Docker image. You'll also learn how to visualize data, including image previews and point clouds, using RViz. Demo based on depthai-ros repository.

Start guide

1. Plug the device

For the best performance please use USB 3.0 port, depend of the camera model. Then use lsusb command to check if the device is visible. The Intel Movidius MyriadX should appear.

2. Clone repository

git clone

3. Select demo folder according to the camera model

cd depthai-docker/demo/<demo_name>
ProductDemo Name

4. Run compose.yaml

xhost local:root
docker compose up


After completing this procedure, RViz should start, containing the RGB image.



The full API of the robot can be found in the official documentation of the device.


This article serves as an excellent introduction to computer vision and AI with OAK cameras within the ROS 2 framework. The demonstration provided by rosbot-xl-telepresence exemplifies the practical application of ROS 2 in the field of mobile robotics.

Do you need any support with completing this project or have any difficulties with software or hardware? Feel free to describe your thoughts on our community forum: or to contact with our support: