A collection of tools used to gather, visualize, and analyze human joint data. Later, this data will train an algorithm to identify human need for help during lifting and carrying tasks.
The code captures human pose data using two methods:
- OptiTrack Motion Capture
- Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Image
The code is compatible with Python2.7 and the following libraries:
The following ROS packages are used (tested with ROS indigo):
For 3D Human Pose Estimation using computer vision, the below repository is included as Git submodule:
From your catkin workspace's src/ directory, clone the repository.
$ git clone --recursive https://github.com/jaydenleong/lift_help_predictor
Setup the repository and install dependencies.
$ cd lift_help_predictor/
$ source setup.sh
To test the code, there is demo data in the lift_help_predictor/data/ directory. To visualize the demo data, enter the following command:
$ roslaunch lift_help_predictor rosbag_visualize.launch
Consult the .md files in lift_help_predictor/docs/ for further documentation and useful instructions about how to use the code.
- Jayden Leong
- Baptiste Busch
- Leonardo Urbano
The following email format may be used: firstname.lastname@epfl.ch
This project is licensed under the terms of the GNU GPLv3 license. By using the software, you are agreeing to the terms of the license agreement.