Human face detection and tracking is a necessary step in many face analysis tasks like face recognition, natural HCI systems, model-based video coding, and content-aware video compression. Although these problems are fairly easy for a human vision system, the machine vision labels them as “hard”.

Our method uses skin color distribution modeling and robust color-based face tracking module to achieve good robustness for low resultion and medium-to-low quality videos. Using skin color as the main tracking cue allows to build an implementation, with low processor load at real-time framerates.

Tracker initialization is performed by Viola & Jones face detection algorithm, implemented in Intel OpenCV.

Examples

Face tracking example videos

Publications

• Vezhnevets V., Sazonov V., Andreeva A., "A Survey on Pixel-Based Skin Color Detection Techniques". Proc. Graphicon-2003, pp. 85-92, Moscow, Russia, September 2003.
  .pdf(93kb) .ps.gz(63kb)
  
  Included in CV-Online: On-Line Compendium of Computer Vision. (Remote copy)
  
  
• V. Vezhnevets "Face and facial feature tracking for natural Human-Computer Interface" Proc. Graphicon - 2002, pp. 86-90, September 16 - September 21, 2002, Nizhny Novgorod, Russia.
  .pdf (513kb) .ps.zg (276kb)
• V. Vezhnevets "Method For Localization Of Human Faces In Color-Based Face Detectors And Trackers" In Proc. Third International Conference on Digital Information Processing And Control In Extreme Situations, pp. 51-56, Minsk, Belarus, May 2002.
  .pdf (83kb) .ps.gz (123kb)

The project team

Contacts