얼굴인식에서의 고유값 조정을 통한 비선형 판별 분석의 향상

Eigenvalue Regularization for Improving Nonlinear LDA in Face Recognition

  • 김상기 (연세대학교 전기전자공학과) ;
  • 이효빈 (연세대학교 전기전자공학과) ;
  • 김성완 (연세대학교 전기전자공학과) ;
  • 이상윤 (연세대학교 전기전자공학과)
  • Kim, Sang-Ki (School of Electrical and Electronic Engineering Yonsei University) ;
  • Lee, Hyo-Bin (School of Electrical and Electronic Engineering Yonsei University) ;
  • Kim, Seong-Wan (School of Electrical and Electronic Engineering Yonsei University) ;
  • Lee, Sang-Youn (School of Electrical and Electronic Engineering Yonsei University)
  • 발행 : 2008.06.18

초록

In this paper, we introduce a novel variant of LDA for face renition. The proposed method is derived by regularizing the eigenvalue of nonlinear LDA. We evaluated the proposed method using AR face database, and it showed outstanding and stable performance over the preceding LDA variants.

키워드