From Matteo Munaro via ros-users@
I would like to introduce OpenPTrack,
an open source project for real time people tracking in RGB-D camera
networks targeted towards applications in education, art, and culture
(e.g. the Whorl installation,
LASHP Welcome Center prototype, and
this education research project.)
OpenPTrack is a joint effort of UCLA REMAP and Open Perception. Key collaborators include the University of Padova, Electroland and Indian
It consists of a collection of ROS packages whose main features are:
- user-friendly network calibration
- person detection from RGB/infrared/depth images
- efficient multi-person tracking
- UDP and NDN streaming of tracking data in JSON format
It currently supports networks of Kinect v1, SwissRanger SR4500 and
stereo cameras custom-built with PointGrey cameras. Support for Kinect v2 will be added soon.
We have three deployed multi-imager testbeds at the UCLA Lab School, UCLA Interpretive Media Laboratory, and University of Indiana, and will add a fourth this month.
We are currently looking for developers interested in creating easy to use front-end interfaces for OpenPTrack, thus fully automating all of the steps of designing the camera network and setting up, configuring, calibrating, operating, and debugging the system.
* Required Skills: *
- Experience with C++.
- Experience with Robot Operating System (ROS).
* Nice To Have: *
- Experience developing web interfaces.
- Experience with ROS tools for developing web interfaces.
- Experience with OpenCV and PCL.
- Knowledge of people tracking algorithms.
- Knowledge of camera calibration algorithms.
In February, we are planning to launch a pilot
crowdfunding campaign to help support getting OpenPTrack in the hands of
more potential users, and would love to get these additional developers
aboard by that time.
For applications and any enquiries, please send an email to Jeff Burke (jburke@remap.ucla.edu) or Matteo Munaro (munaro@dei.unipd.it).
Leave a comment