Face-Sketch Recognition Using Canonical Correlation Analysis

Sen B., Ozkazanc Y.

21st Signal Processing and Communications Applications Conference (SIU), CYPRUS, 24 - 26 April 2013, (Full Text)

  • Publication Type: Conference Paper / Full Text
  • Volume:
  • Doi Number: 10.1109/siu.2013.6531277
  • Country: CYPRUS
  • Hacettepe University Affiliated: Yes

Abstract

Hand-drawn face sketches are frequently used in criminal investigations. In this paper, we present a novel framework for face recognition from sketches. Our framework based is on Principle Component Analysis (PCA) and Canonical Correlation Analysis (CCA). First, we apply PCA to a dataset for dimension reduction and then apply CCA for reaching maximum correlation within a dataset. This approach is tested on two different datasets including 311 photo-sketch pairs. The performance reached 99.36% recognition rate on these experiments.