A new constrained nonnegative matrix factorization for facial expression recognition

Viet Hang Duong, Manh Quan Bui, Pham The Bao, Jia Ching Wang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A new NMF model, spatial constrained graph sparse nonnegative matrix factorization (SGSNMF) is adopted for facial expression recognition. In this model, the extracted features preserve the topological structure of the original images and achieve sparseness from L2 constraint on coefficient matrix, meanwhile the base satisfy pixel dispersion penalty. The proposed method takes advantage of the project gradient decent and is based on the alternating nonnegative least square framework. Experiments on two facial expression recognition scenarios that involve a whole face and an occluded face reveal that the proposed algorithm outperforms the prevalent NMF methods.

Original languageEnglish
Title of host publicationProceedings of the 2017 International Conference on Orange Technologies, ICOT 2017
EditorsLei Wang, Minghui Dong, Yanfeng Lu, Haizhou Li
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages79-82
Number of pages4
ISBN (Electronic)9781538632758
DOIs
StatePublished - 10 Apr 2018
Event5th International Conference on Orange Technologies, ICOT 2017 - Singapore, Singapore
Duration: 8 Dec 201710 Dec 2017

Publication series

NameProceedings of the 2017 International Conference on Orange Technologies, ICOT 2017
Volume2018-January

Conference

Conference5th International Conference on Orange Technologies, ICOT 2017
Country/TerritorySingapore
CitySingapore
Period8/12/1710/12/17

Keywords

  • Facial expression recognition
  • Graph regularization
  • Nonnegative matrix factorization
  • Projectedgradient descent

Fingerprint

Dive into the research topics of 'A new constrained nonnegative matrix factorization for facial expression recognition'. Together they form a unique fingerprint.

Cite this