Pose classification of human faces by weighting mask function approach

Chiunhsiun Lin, Kuo Chin Fan

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

This paper presents a novel method for automatic estimation of the poses/degrees of human faces. The proposed system consists of two main parts. The first part is the searching of potential face regions that are gotten from the isosceles-triangle criteria based on the rules of "the combination of two eyes and one mouth". The second part of the proposed system is the performing the task of pose verification by utilitizing face weighting mask function, direction weighting mask function, and pose weighting mask function. The proposed face poses/degrees classification system can also determine the poses of multiple faces. Experimental results demonstrate that an approximately 99% success rate is achieved and the relative false estimation rate is very low.

Original languageEnglish
Pages (from-to)1857-1869
Number of pages13
JournalPattern Recognition Letters
Volume24
Issue number12
DOIs
StatePublished - Aug 2003

Keywords

  • Face detection
  • Face poses classification
  • Triangle-based criteria
  • Weighting mask function

Fingerprint

Dive into the research topics of 'Pose classification of human faces by weighting mask function approach'. Together they form a unique fingerprint.

Cite this