RTFTR logo

 docs | downloads (stat) | faq | trac | wiki | dev. mailing list | svn repo. | awards | screenshots

Project Description:

Real time face tracking and recognition refers to the task of locating human faces in a video stream and identifying the faces by matching them against the database of known faces. There are two novel techniques that are being tested in RTFTR:

  1. Study the effect of using two or more face extraction/recognition algorithms, in parallel, on overall accuracy of tracking/recognition task
  2. Use High Performance Distributed Computing ( using OpenMPI and OpenMP ) to achieve real time capability for tracking/recognition task

We are applying a modular approach to development of face recognition systems which allows users to form a custom processing pathway for the task of face extraction/recognition. A flexible 6 stage processing pathway is being used in RTFTR. Two stages are dedicated for the task of face extraction and recognition. RTFTR will have option to chose from a wide variety of sources (like a CCD camera, video sequence, images, etc).

We are implementing the following four algorithms (two for each face extraction/recognition task) :

We have completed the development of RTFTR version 0.1 (which was submitted to Dept. of ECE, IOE-Pulchowk Campus). We participated with RTFTR version 0.2 in OSCN-2009.

Project News:

Documents:

Demo Videos:


Demo video for OSCN-2009


Demo video for IOE

Developers:

Abhishek Dutta Anjan Nepal Bibek Shrestha Lakesh Kansakar

(Left to Right): Lakesh, Bibek, Anjan, Abhishek (Sep. 10, 2008)

SourceForge.net Logo

Last Updated: May. 11, 2012