DocumentCode :
615156
Title :
A comparison of alternative classifiers for detecting occurrence and intensity in spontaneous facial expression of infants with their mothers
Author :
Zaker, N. ; Mahoor, M.H. ; Mattson, Whitney I. ; Messinger, D.S. ; Cohn, J.F.
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Denver, Denver, CO, USA
fYear :
2013
fDate :
22-26 April 2013
Firstpage :
1
Lastpage :
6
Abstract :
To model the dynamics of social interaction, it is necessary both to detect specific Action Units (AUs) and variation in their intensity and coordination over time. An automated method that performs well when detecting occurrence may or may not perform well for intensity measurements. We compared two dimensionality reduction approaches - Principal Components Analysis with Large Margin Nearest Neighbor (PCA+LMNN) and Laplacian Eigenmap - and two classifiers, SVM and K-Nearest Neighbor. Twelve infants were video-recorded during face-to-face interactions with their mothers. AUs related to positive and negative affect were manually coded from the video by certified FACS coders. Facial features were tracked using Active Appearance Models (AAM) and registered to a canonical view before extracting Histogram of Oriented Gradients (HOG) features. All possible combinations of dimensionality reduction approaches and classifiers were tested using a leave-onesubject-out cross-validation. For detecting consistency (i.e. reliability as measured by ICC), PCA+LMNN and SVM classifiers gave best results.
Keywords :
eigenvalues and eigenfunctions; face recognition; gradient methods; object detection; pattern classification; principal component analysis; support vector machines; AAM; FACS coder; HOG feature; LMNN; Laplacian eigenmap; PCA; SVM; action unit; active appearance model; dimensionality reduction; face-to-face interaction; facial feature; histogram of oriented gradient feature; infant; k-nearest neighbor; large margin nearest neighbor classifier; leave-one-subject-out cross-validation; mother; principal components analysis; social interaction dynamics; spontaneous facial expression; Face; Face recognition; Feature extraction; Gold; Histograms; Principal component analysis; Support vector machines; Action Unit; Facial Expressions; Histogram of Oriented Gradients; Laplacian Eigenmap; Structural SVM;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Automatic Face and Gesture Recognition (FG), 2013 10th IEEE International Conference and Workshops on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4673-5545-2
Electronic_ISBN :
978-1-4673-5544-5
Type :
conf
DOI :
10.1109/FG.2013.6553795
Filename :
6553795
Link To Document :
بازگشت