📦 Alterego Object Detection

Alterego Object Detection is a ROS-compatible object detection package that leverages YOLOv8 (Ultralytics) to identify objects in real time. In addition to standard object detection, the package extracts the 3D relative position of the detected objects with respect to an Intel RealSense camera.

✨ Features

• Real-time object detection using YOLOv8 (Ultralytics)
• Extraction of 3D coordinates (X, Y, Z) of detected objects relative to the camera
• Support for Intel RealSense RGB-D cameras
• ROS integration for easy use in robotic applications

📋 Requirements

• Python 3
• ROS Noetic
• Ultralytics YOLOv8
• Intel RealSense SDK (pyrealsense2)
• OpenCV
• NumPy

You can install the required Python packages directly with create_env.sh

🛠️ Clone the package

cd ~/(your_catkin_path)/src/AlterEgo_v2/utils/alterego_object_dection
git clone https://github.com/IIT-SoftBots/alterego_object_detection.git

🛠️ Create conda env

These four commands download the latest 64-bit version of the Linux installer, rename it to a shorter file name, silently install, and then delete the installer:

mkdir -p ~/miniconda3
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda3/miniconda.sh
bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3
rm ~/miniconda3/miniconda.sh

ATTENTION: After installing, close and reopen your terminal application or refresh it by running the following command:

source ~/miniconda3/bin/activate

To initialize conda on the current shell , run the following command:

conda init

Then deactivate the autostart activation of the base env

conda config --set auto_activate_base false

⚙️ Build the entire solution

If you want to use the object detection, lets create the virtual env from the bash file create_env.sh inside .

(optional) Be sure to accept Conda's ToS on first Conda use:

conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/main
conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/r

Create Conda env and build packages:

cd ~/(your_catkin_path)/src/AlterEgo_v2/utils/alterego_object_dection
chmod +x create_env.sh # To make executable the script
./create_env.sh
cd ~/(your_catkin_path)
catkin build alterego_object_detection

Move to source dir cd ~/(your_catkin_path)/. Then

catkin build
conda deactivate

🚀 How to use

Run the object_detection.py file as you prefer:

roslaunch alterego_object_detection object_detection.launch 

or

rosrun alterego_object_detection object_detection.py 

The robot start to classify the detected object and to estimate their relative position with respect to the realsense.

If you are interested to focus on a single object class (for istance bottle), we have to public in the related topic:

rostopic pub $(robot_name)$/desired_object std_msgs/String "data: 'bottle'"

N.B.: $(robot_name)$ is the env parameter, for istances robot_alterego5 or robot_alterego_sim.

🧑‍💻 Contributors

Samuele Bordini
      Email: samuele.bordini@phd.unipi.it
      GitHub: @SamueleBordini
      Linkedin: @samuele-bordini-linkedin

Do Won Park
      Email: do.park@iit.it
      GitHub: @tonappa
      Linkedin: @dowon-park-linkedin