Motion Detection and Tracking

• Project: Control with Motion in the Loop

  This project will focus on the setup of the basic hardware and on the development of the basic software necessary for a simple motion tracking experiment. Our experimental tracking system is based on a camera mounted on a Pan-Tilt unit (``head''), which provides positioning capabilities. The application example we are aiming at is the following: a person enters a room and moves within the camera field of view; the vision based control system has to track the person's movement.

  We plan to start with the following simplified experiment: a white ball moves on a black background while a threshold algorithm provides ``position'' informations to a simplified feedback control law. Once this basic experiment is successfully implemented, more sophisticated tracking systems will become subjects of small additional projects.

  We used to use TI C80 DSP chips for image capturing and algorithm implementation. For simplicity and better performance, from 1999 we are going to use fast Pentium machine with a framegrabber and wide-angle camera.

  • Hardware: Pentium II, Motorized Zoom CCD camera, Pan-Tilt Head unit

  • Software: C++.

• Bibliography
  • M.J. Swain and D.H. Balard, "Color Indexing", International Journal in Computer Vision, Vol 7, No 1, 1991.

  • N. Papanikolopoulos, P. Khosla, T. Kanade, ``Visual tracking of a moving target by a camera mounted on a robot: a combination of control and vision'', IEEE Trans. Robotics and Automation, Vol 9, No 1, February 1993.

  • D. Koller, J. Weber and J. Malik, "Robust Multiple Car Tracking with Occlusion Reasoning", Technical Report UCB/CSD 93-780, Computer Science Division, UC Berkeley, 1993.

• Handouts:
  • ProjectHandout (postcript)

• Students Reports:
  • students report 95 (postcript)

Back to the home page for CNS/EE248