Advertisement

Automatic Eye Position Detection and Tracking Under Natural Facial Movement

  • Yan Tong
  • Qiang Ji
Part of the Signals and Commmunication Technologies book series (SCT)

Abstract

Automatic and precise detection and tracking of facial features, and in particular the eyes, are important in many applications of Passive Eye Monitoring including Driver Fatigue Detection, Cognitive Driver Distraction and Gaze- Based Interaction. Generally, the facial feature tracking technologies could be classified into two categories: model-free and model-based tracking algorithms. The model-free tracking algorithms [65,217,220,229,384,497,563,624, 627, 720] are general purpose point trackers without the prior knowledge of the object. Each facial feature point is usually tracked by performing a local search for the best matching position, around which the appearance is most similar to the one in the initial frame. However, the model-free methods are susceptible to the inevitable tracking errors due to the aperture problems, noise, and occlusion. Model-based methods, on the other hand, focus on explicit modeling the shape of the objects. Recently, extensive work has been focused on the shape representation of deformable objects such as active contour models (Snakes) [314], deformable template method [708], active shape model (ASM) [105], active appearance model (AAM) [102], direct appearance model (DAM) [266], elastic bunch graph matching (EBGM) [682], morphable models [305], and active blobs [558]. Although the model-based methods utilize much knowledge on the face to realize an effective tracking, these models are limited to some common assumptions, e.g. a nearly frontal view face and moderate facial expression changes, and tend to fail under large pose variations or facial deformations in real world applications.

Keywords

Feature Point Facial Feature Shape Model Active Appearance Model Active Shape Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Yan Tong
    • 1
  • Qiang Ji
    • 2
  1. 1.Rensselaer Polytechnic InstituteTroyUSA
  2. 2.Department of Electrical ComputerRensselaer Polytechnic InstituteTroyUSA

Personalised recommendations