首页|Topological principal component analysis for face encoding and recognition

Topological principal component analysis for face encoding and recognition

扫码查看
原文链接
  • NETL
  • NSTL
Principal component analysis (PCA)-like methods make use of an estimation of the covariances between sample variables. This estimation does not take into account their topological relationships. This paper proposes how to use these relationships in order to estimate the covariances in a more robust way. The new method topological principal component analysis (TPCA) is tested using both face encoding and recognition experiments showing how the gener- alization capabilities of PCA are improved.

GeneralizationPrincipal component analysisFace recognitionTopological covariance matrixCovariance estimation

Albert Pujol、Jordi Vitria、Felipe Lumbreras

展开 >

2001

Pattern recognition letters

Pattern recognition letters

EI
ISSN：0167-8655
年,卷(期)：2001.22(6/7)