ros_vision_track

Directory Structure

├── ros_vision_track
│   ├── config
│   │   └── rviz
│   ├── ros_vision_track
│   │   └── camera_processing
│   │       ├── trackers
│   │       ├── weights
│   │       └── yolov5
│   ├── launch
│   ├── resource
│   └── test

Note: This is setup like a ROS2 package, but you do not need to build it to run the scripts. However to use launch files the package needs to be built.

Setup

To get started

git clone --recurse-submodules https://github.com/adrian-soch/ros_vision_track.git

Install Python requirements

# Pytorch GPU Version go here: https://pytorch.org/get-started/locally/

# (Optional) PyTorch CPU Version
pip install torch torchvision

pip install numpy

# OpenCV CPU Version
pip install opencv-python
pip install opencv-contrib-python

Yolov5 requirements

numpy>=1.18.5
opencv-python>=4.1.1
Pillow>=7.1.2
psutil  # system resources
PyYAML>=5.3.1
requests>=2.23.0
scipy>=1.4.1

Running the Code

Image detection and tracking:

# Play the rosbag
ros2 bag play <Path to rosbag>/<bag name>.db3

# Start the python code
cd <Path to ros_vision_track>/ros_vision_track
python3 camera_processing_yolovX_node.py

Modifying parameters

To modify the parameters go to ros_vision_track/camera_processing/vision_track.py.

References

1. https://github.com/ultralytics/yolov5
2. https://github.com/WongKinYiu/yolov7
3. https://github.com/mikel-brostrom/Yolov5_StrongSORT_OSNet