🦾 Skeleton Trajectory Repository

Welcome to the Skeleton Trajectory Repository project of the CARI with Vis4Mechs Lab

This repository develops a ROS-based framework to identify human keypoints and reconstruct them in 3D within a collaborative robotic cell, enabling effective human movement tracking.

🚀 Start the Skeleton Tracking Node

The entire framework can be run in a simplified way via a launcher file that automatically handles the different nodes:

roslaunch skeleton_trajectory skeleton_trajectory.launch

⚙️ Launch Parameters

The following parameters are available:

• kalmanBase: specifies the filter type.
  • kalmanBase:=True: the default condition, filters the points independently (assuming constant acceleration in Cartesian space).
  • kalmanBase:=False: applies filtering based on a model of the person’s limb kinematics (assuming constant acceleration in joint space).

📋 Requirements

• Create a venv:

  python -m venv /path/to/new/virtual/environment

• Install requirements (pip or pip3) (check path to requirements):

  pip3 install -r requirements.txt

• Activate the virtual environment:

  source path_to_venv/bin/activate

👥 Maintainers

• Samuele Sandrini, SamueleSandrini
• Manuel Beschi, ManuelBeschi