object-detection

OpenVINO Vision Applications

Overview

This repository contains multiple vision applications that leverage OpenVINO for computer vision tasks in robotics, including object detection, semantic segmentation, and YOLOv8-based detection. These applications provide ROS 2 integration for real-time perception capabilities using Intel RealSense cameras and OpenVINO inference engine.

Get Started

System Requirements

Prepare the target system following the official documentation.

Build

To build debian packages, export ROS_DISTRO env variable to desired platform and run make package command. After build process successfully finishes, built packages will be available in the root directory. The following command is an example for Jazzy distribution.

ROS_DISTRO=jazzy make package

You can list all built packages:

$ ls | grep -i .deb
ros-jazzy-object-detection-tutorial_2.3-1_amd64.deb
ros-jazzy-segmentation-realsense-tutorial_2.3-1_amd64.deb
ros-jazzy-yolo-msgs_2.3-1_amd64.deb
ros-jazzy-yolo_2.3-1_amd64.deb

To clean all build artifacts:

make clean

Install

If Ubuntu 22.04 with Humble is used, then run

source /opt/ros/humble/setup.bash

If Ubuntu 24.04 with Jazzy is used, then run

source /opt/ros/jazzy/setup.bash

Finally, install the Debian packages that were built via make package:

sudo apt update
sudo apt install ./ros-$(ROS_DISTRO)-object-detection-tutorial_*_amd64.deb
sudo apt install ./ros-$(ROS_DISTRO)-segmentation-realsense-tutorial_*_amd64.deb
sudo apt install ./ros-$(ROS_DISTRO)-yolo-msgs_*_amd64.deb
sudo apt install ./ros-$(ROS_DISTRO)-yolo_*_amd64.deb

Test

To run unit tests (implemented with colcon) execute the below command with target ROS_DISTRO (example for Jazzy):

ROS_DISTRO=jazzy make test

Development

There is a set of prepared Makefile targets to speed up the development.

In particular, use the following Makefile target to run code linters.

make lint

To run license compliance validation:

make license-check

To see a full list of available Makefile targets:

$ make
Target               Description
------               -----------
clean                Clean up all build artifacts
license-check        Perform a REUSE license check using docker container https://hub.docker.com/r/fsfe/reuse
lint                 Run all sub-linters using super-linter (using linters defined for this repo only)
lint-all             Run super-linter over entire repository (auto-detects code to lint)
package              Build Debian packages
source-package       Create source package tarball
test                 Test code using colcon

Usage

This repository contains three main applications:

Object Detection Application

The Object Detection application provides real-time object detection capabilities using OpenVINO and Intel RealSense cameras. It supports various pre-trained models and can be customized for specific detection tasks.

For detailed usage instructions, see the Object Detection Tutorial documentation.

Segmentation Realsense Tutorial

The Segmentation application demonstrates semantic segmentation using OpenVINO with Intel RealSense depth cameras. It provides pixel-level classification for scene understanding.

For detailed usage instructions, see the Segmentation Realsense Tutorial documentation.

YOLOv8

The YOLOv8 application provides state-of-the-art object detection using the YOLOv8 model optimized with OpenVINO. It offers high-performance real-time detection for robotics applications.

For detailed usage instructions, see the YOLOv8 OpenVINO Tutorial documentation.

License

OpenVINO Vision Applications is licensed under Apache 2.0 License.