• Human and Object Detection with Intel D435 Depth Camera
  • Table of contents
  • 1. Introduction
  • 2. Installation guide
    • 2.1. System setup
    • 2.2. Installing dependencies
    • 2.3. Building project
  • 3. Launching project

Human and object detection with Intel D435 Depth Camera is applied by using darknet_ros package.

• https://github.com/leggedrobotics/darknet_ros

Setup is done according to the video in the following youtube link with the source code in the drive link.

• https://www.youtube.com/watch?v=jxH68h8wzTM
• https://drive.google.com/drive/folders/1UHXizZZQEcFSx9R2IreEWMQO6SBcUyPy

This project is developed for Ubuntu 20.04 with ROS Noetic. Be sure your system has the same configuration. You can check the following links to install them:

• Ubuntu 20.04 setup guide link: https://releases.ubuntu.com/focal/
• ROS Noetic setup guide link: https://wiki.ros.org/noetic/Installation/Ubuntu

It is required that the ROS Noetic should be sourced in each terminal to run the simulation as mentioned in the ROS Noetic setup guide link. To do that,

source /opt/ros/noetic/setup.bash

should be sourced in each terminal. To make it easier, it can be added to the .bashrc file for automatic sourcing. Details can be found in the ROS Noetic setup guide link. Sourcing ROS Noetic will no longer be mentioned in further steps.

Bonus: If you have a different Linux or ROS distro than the specified requirements, you may prefer to use a Docker container. To do that, check the docker folder.

To have a clear installation, updating and upgrading your system is recommended.

sudo apt-get update
sudo apt-get upgrade

For the installation of packages, basic tools are required.

sudo apt install git apt-utils python3-catkin-tools wget libtool

Basic ROS packages are required for the project.

sudo apt install ros-noetic-cv-bridge ros-noetic-image-transport ros-noetic-tf ros-noetic-ddynamic-reconfigure ros-noetic-diagnostic-updater

Realsense2 is required for the Intel D435 Camera.

sudo apt install ros-noetic-realsense2-camera ros-noetic-realsense2-description ros-noetic-rgbd-launch

Check whether OpenCV version is greater than 4, i.e. OpenCV4 is installed.

pkg-config --modversion opencv4

If not, install OpenCV4 by following the link below

• https://rodosingh.medium.com/using-cmake-to-build-and-install-opencv-for-python-and-c-in-ubuntu-20-04-6c5881eebd9a

Then, go to opt/ros/noetic/share/cv_bridge/cmake and change the cv_bridgeConfig.cmake file: Find line set(_include_dirs "include;/usr/include;/usr/include/opencv") and change with set(_include_dirs "include;/usr/include;/usr/local/include/opencv4")

Open a terminal (will be mentioned as T1) and create a directory preferably in the home location as

mkdir -p sabes_ws
cd sabes_ws

Clone the project repository (T1)

git clone [email protected]:cemkucukgenc/darknet_ros_3D_v1.git

To build the project in sabes_ws/darknet_ros_3D_v1/catkin_ws (T1)

cd darknet_ros_3D_v1/catkin_ws
catkin build

To run the project (T1)

source /sabes_ws/darknet_ros_3D_v1/catkin_ws/devel/setup.bash
roslaunch launch_inference launch_inference.launch

To see the output in the ROS topic (T2)

source /sabes_ws/darknet_ros_3D_v1/catkin_ws/devel/setup.bash
rostopic echo /objects_position/message

Bonus: Bookmarked issue pages that helped me to debug

• leggedrobotics/darknet_ros#381
• leggedrobotics/darknet_ros#25
• https://stackoverflow.com/questions/55605052/why-is-the-output-video-too-slow-in-darknet
• pjreddie/darknet#80
• IntelRealSense/realsense-ros#1187