Scale Normalization for the Distance Maps AAM - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2006

Scale Normalization for the Distance Maps AAM

Résumé

The Active Apearence Models (AAM) are often used in Man-Machine Interaction for their ability to align the faces. We propose a new normalization method for AAM based on distance map in order to strengthen their robustness to differences in illumination. Our normalization do not use the photometric normalization protocol classically used in AAM and is much more simpler to implement. Compared to Distance Map AAM performances of {Leg06} and other AAM implementation which use CLAHE {Zuiderveld94} normalization or gradient information, our proposition is at the same time much robust to illumination and AAM initialization. The tests have been drive in the context of generalization: 10 persons with frontal illumination from M2VTS database {m2vts} were considered to build the AAM, and 17 persons under 21 different illuminations from CMU database {Sim02} were used for the testing base.
Fichier principal
Vignette du fichier
ICARCV2006.pdf (682.63 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00143460 , version 1 (25-04-2007)

Identifiants

  • HAL Id : hal-00143460 , version 1

Citer

Denis Giri, Maxime Rosenwald, Benjamin Villeneuve, Sylvain Le Gallou, Renaud Seguier. Scale Normalization for the Distance Maps AAM. International Conference on Control, Automation, Robotics and Vision (ICARCV), Dec 2006, Singapore, Singapore. pp.100. ⟨hal-00143460⟩
158 Consultations
176 Téléchargements

Partager

Gmail Facebook X LinkedIn More