Analysis of landmarks in recognition of face expressions

N. Alugupally, A. Samal, D. Marx, S. Bhatia

Research output: Contribution to journalArticle

7 Scopus citations

Abstract

Facial expression is a powerful mechanism used by humans to communicate their emotions, intentions, and opinions to each other. The recognition of facial expressions is extremely important for a responsive and socially interactive human-computer interface. Such an interface with a robust capability to recognize human facial expressions should enable an automated system to effectively deploy in a variety of applications, including human computer interaction, security, law enforcement, psychiatry, and education. In this paper, we examine several core problems in face expression analysis from the perspective of landmarks and distances between them using a statistical approach. We have used statistical analysis to determine the landmarks and features that are best suited to recognize the expressions in a face. We have used a standard database to examine the effectiveness of landmark based approach to classify an expression (a) when a face with a neutral expression is available, and (b) when there is no a priori information about the face.

Original languageEnglish (US)
Pages (from-to)681-693
Number of pages13
JournalPattern Recognition and Image Analysis
Volume21
Issue number4
DOIs
StatePublished - Dec 2011

Keywords

  • Face expressions
  • face features
  • linear discriminant analysis
  • statistical analysis

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Computer Graphics and Computer-Aided Design

Fingerprint Dive into the research topics of 'Analysis of landmarks in recognition of face expressions'. Together they form a unique fingerprint.

  • Cite this