Facial feature localization using MOSSE correlation filters

David S. Bolme, J. Ross Beveridge

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

2 Scopus citations

Abstract

Accurately measuring the location of facial features is an important step in many face recognition algorithms. Every face is unique which means localization needs to be tolerant of differences between individual subjects. Additionally, changing illumination, poor focus, and deformation due to expression changes complicate the problem. This paper introduces a method for locating facial features that uses Minimum Output Sum of Squared Error (MOSSE) correlation filters to model object appearance and is combined with a Robust Active Shape Model (ASM) to model facial geometry. It is demonstrated that MOSSE correlation filters outperform Stasm (an open source ASM implementation), Gabor Jets and in some cases even matches human performance.

Original languageEnglish
Title of host publicationFIIW 2012 - 2012 Future of Instrumentation International Workshop Proceedings
Pages124-127
Number of pages4
DOIs
StatePublished - 2012
Event2012 Future of Instrumentation International Workshop, FIIW 2012 - Gatlinburg, TN, United States
Duration: Oct 8 2012Oct 9 2012

Publication series

NameFIIW 2012 - 2012 Future of Instrumentation International Workshop Proceedings

Conference

Conference2012 Future of Instrumentation International Workshop, FIIW 2012
Country/TerritoryUnited States
CityGatlinburg, TN
Period10/8/1210/9/12

Keywords

  • Biometrics
  • Face Recognition
  • Landmark Localization

Fingerprint

Dive into the research topics of 'Facial feature localization using MOSSE correlation filters'. Together they form a unique fingerprint.

Cite this