Multiple faces tracking using motion prediction and IPCA in particle filters

  • Authors:

  • Sukwon Choi;Daijin Kim

  • Affiliations:

  • Dept. of Computer Science & Engineering, Pohang University of Science and Technology, Pohang, Korea;Dept. of Computer Science & Engineering, Pohang University of Science and Technology, Pohang, Korea

  • Venue:
  • ICB'07 Proceedings of the 2007 international conference on Advances in Biometrics
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose an efficient real-time face tracking system that can track fast moving face and cope with the illumination changes. To achieve these goals, we use the active appearance model(AAM) to represent the face image due to its simplicity and flexibility and take the particle filter framework to track the face image due to its robustness. We modify the particle filter framework as follows. To track fast moving face, we predict the motions using motion history and motion estimation, hence we can reduce the required number of particles. For observation model, we use active appearance model(AAM) to obtain an accurate face region, and update the model using incremental principle component analysis(IPCA). Occlusion handling scheme incorporates motion history to handle the moving face with occlusion. We have expanded our application to multiple faces tracking system. Experimental results present the robustness and effectiveness of the proposed system.