Local face sketch synthesis learning

  • Authors:
  • Xinbo Gao;Juanjuan Zhong;Dacheng Tao;Xuelong Li

  • Affiliations:
  • School of Electronic Engineering, Xidian University, Xi'an, Shaanxi, PR China;School of Electronic Engineering, Xidian University, Xi'an, Shaanxi, PR China;Biometrics Research Centre, The Hong Kong Polytechnic University, Hong Kong, PR China;School of Computer Science and Information Systems, Birkbeck College, University of London, UK

  • Venue:
  • Neurocomputing
  • Year:
  • 2008

Quantified Score

Hi-index 0.01

Visualization

Abstract

Facial sketch synthesis (FSS) is crucial in sketch-based face recognition. This paper proposes an automatic FSS algorithm with local strategy based on embedded hidden Markov model (E-HMM) and selective ensemble (SE). By using E-HMM to model the nonlinear relationship between a photo-sketch patch pair, a series of pseudo-sketch patches, generated based on several learned models for a given photo patch, are integrated with SE strategy to synthesize a finer face pseudo-sketch patch. Finally, the intact pseudo-sketch can be generated by combining all synthesized patches. Experimental results illustrate that the proposed FSS algorithm works well.