Unsupervised discriminant projection analysis for feature extraction

Jian Yang, Dapeng Zhang, Zhong Jin, Jing Yu Yang

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

21 Citations (Scopus)


This paper develops an unsupervised discriminant projection (UDP) technique for feature extraction. UDP takes the local and non-local information into account, seeking to find a projection that maximizes the non-local scatter and minimizes the local scatter simultaneously. This characteristic makes UDP more intuitive and more powerful than the up-to-date method - Locality preserving projection (LPP, which considers the local information only) for classification tasks. The proposed method is applied to face biometrics and examined using the ORL and FERET face image databases. Our experimental results show that UDP consistently outperforms LPP, PCA, and LDA.
Original languageEnglish
Title of host publicationProceedings - 18th International Conference on Pattern Recognition, ICPR 2006
Number of pages4
Publication statusPublished - 1 Dec 2006
Event18th International Conference on Pattern Recognition, ICPR 2006 - Hong Kong, Hong Kong
Duration: 20 Aug 200624 Aug 2006


Conference18th International Conference on Pattern Recognition, ICPR 2006
Country/TerritoryHong Kong
CityHong Kong

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this