Sciweavers

Share
CIARP
2008
Springer

Ensemble Approaches to Facial Action Unit Classification

10 years 14 days ago
Ensemble Approaches to Facial Action Unit Classification
Facial action unit (au) classification is an approach to face expression recognition that decouples the recognition of expression from individual actions. In this paper, upper face aus are classified using an ensemble of MLP (Multi-layer perceptron) base classifiers with feature ranking based on PCA components. This approach is compared experimentally with other popular feature-ranking methods applied to Gabor features. Experimental results on Cohn-Kanade database demonstrate that the MLP ensemble is relatively insensitive to the feature-ranking method but optimized PCA features achieve lowest error rate. When posed as a multi-class problem using ErrorCorrecting-Output-Coding (ECOC), error rates are comparable to two-class problems (one-versus-rest) when the number of features and base classifier are optimized.
Terry Windeatt, Kaushala Dias
Added 12 Oct 2010
Updated 12 Oct 2010
Type Conference
Year 2008
Where CIARP
Authors Terry Windeatt, Kaushala Dias
Comments (0)
books