Multifocal System for Real Time Tracking of Dynamic Facial and Body Features

Funding agency: Romanian National Authority for Scientific Research, CNDI–UEFISCDI

Project type:Joint Applied Research Project, PN-II-RU-TE-2014-4-1746.

Contract number: 6 / 2015
Project coordinated by the Technical University of Cluj-Napoca



Human interaction occurs on multiple communication channels: speech, intonation (paralanguage), facial expressions and body postures. The project’s general objective is to design sensorial systems, models, detection and measurement algorithms, for tracking the head, eyes and facial features, able to work indoor and outdoor, without constraints on the user behavior. Due to our extensive experience with stereovision, which is a sensor for both near and far distances, and is able to deliver both range and image data, the solution will be stereovision-centered. As the stereovision system is computationally intensive, it must be limited in image resolution and frame rate, and for this reason we propose that we circumvent these limitations by adding a third camera to the rig, a high speed camera of a narrower field of view, oriented by a pan tilt unit to have the observed face in focus. The project’s results are relevant to all fields that are interested in detailed observation of individuals: psychology, psychiatry, education, driving attention monitoring, biometrics, etc. The project’s results will be a collection of algorithms, models, techniques, hardware systems and image and feature databases which will be of great interest to any researcher in the field.