FACIAL EXPRESSION RECOGNITION FROM STATIC IMAGES

Akkoca B. S., Gökmen M.

22nd IEEE Signal Processing and Communications Applications Conference (SIU), Trabzon, Turkey, 23 - 25 April 2014, pp.1291-1294 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/siu.2014.6830473
  • City: Trabzon
  • Country: Turkey
  • Page Numbers: pp.1291-1294
  • Istanbul Technical University Affiliated: Yes

Abstract

In this study, a cascaded structure includes algorithms of Local Zernike Moments, Local XOR Patterns and Global Zernike Moments is used for Facial Expression Recognition. 10 patches are specified on the images in CK+ dataset and over those pathces 3 methods are applied respectively. By concataneting of the moment coefficients, the feature vector generated and classified with Support Vector Machines.

In this study, a cascaded structure includes algorithms of Local Zernike Moments, Local XOR Patterns and Global Zernike Moments is used for Facial Expression Recognition. 10 patches are specified on the images in CK+ dataset and over those pathces 3 methods are applied respectively. By concataneting of the moment coefficients, the feature vector generated and classified with Support Vector Machines.