Face identification performance using facial expressions as perturbation

Authors:
Minoru Nakayama;Takashi Kumakura
Affiliations:
CRADLE, Tokyo Institute of Technology, Tokyo, Japan;Human System Science, Tokyo Institute of Technology, Tokyo, Japan
Venue:
ICANN'05 Proceedings of the 15th international conference on Artificial Neural Networks: biological Inspirations - Volume Part I
Year:
2005

Citing 2
Cited 1

Training Invariant Support Vector Machines

Machine Learning
SVMTorch: support vector machines for large-scale regression problems

The Journal of Machine Learning Research

Human facial expression recognition using hybrid network of PCA and RBFN

ICANN'06 Proceedings of the 16th international conference on Artificial Neural Networks - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

The paper presents improvements in face identification performance using synthesized images as a perturbation method. Three facial expression features, smiles, anger and screams, were extracted from images of actual facial expression using the eigenspace method. Synthesized facial images based on these features were added to learning data of a personal identification model using support vector machines (SVM). The performance of this model was significantly higher than that of a model trained without facial expression images, but significantly lower than that of a model using actual expression images. The results suggest that identification performance also depends significantly on facial expression.