Head Pose Classification Using a BidimensionalCorrelation Filter

Type : Publication
Auteur(s) :  Djemel Ziou, Dayron RIZO-RORDRIGUEZ, Antoine Tabbone, Nafaa Nacereddine
Année :  2015
Domaine : Electronique
Revue : Lecture Notes in Computer Science
Résumé en PDF :  (résumé en pdf)
Fulltext en PDF :  (.pdf)
Mots clés :  Head pose classification, Bidimensional correlation filter

Résumé : 

Correlation filters have been extensively used in face recognition but surprisingly underused in head pose classification. In this paper,we present a correlation filter that ensures the tradeoff between three criteria: peak distinctiveness, discrimination power and noise robustness.Such a filter is derived through a variational formulation of these threecriteria. The closed form obtained intrinsically considers multiclass information and preserves the bidimensional structure of the image. The filterproposed is combined with a face image descriptor in order to deal withpose classification problem. It is shown that our approach improves poseclassification accuracy, especially for non-frontal poses, when comparedwith other methods.