Automatic facial landmark tracking in video sequences using kalman filter assisted active shape models

  • Authors:
  • Utsav Prabhu;Keshav Seshadri;Marios Savvides

  • Affiliations:
  • Department of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, PA;Department of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, PA;Department of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, PA

  • Venue:
  • ECCV'10 Proceedings of the 11th European conference on Trends and Topics in Computer Vision - Volume Part I
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper we address the problem of automatically locating the facial landmarks of a single person across frames of a video sequence. We propose two methods that utilize Kalman filter based approaches to assist an Active Shape Model (ASM) in achieving this goal. The use of Kalman filtering not only aids in better initialization of the ASM by predicting landmark locations in the next frame but also helps in refining its search results and hence in producing improved fitting accuracy. We evaluate our tracking methods on frames from three video sequences and quantitatively demonstrate their reliability and accuracy.