Multi-task Learning of Facial Landmarks and Expression

Author

Devries, Terrance ; Biswaranjan, Kumar ; Taylor, Graham W.

Author_Institution

Sch. of Eng., Univ. of Guelph, Guelph, ON, Canada

fYear

2014

fDate

6-9 May 2014

Firstpage

98

Lastpage

103

Abstract

Recently, deep neural networks have been shown to perform competitively on the task of predicting facial expression from images. Trained by gradient-based methods, these networks are amenable to "multi-task" learning via a multiple term objective. In this paper we demonstrate that learning representations to predict the position and shape of facial landmarks can improve expression recognition from images. We show competitive results on two large-scale datasets, the ICML 2013 Facial Expression Recognition challenge, and the Toronto Face Database.

Keywords

face recognition; image representation; learning (artificial intelligence); neural nets; ICML 2013 facial expression recognition challenge; Toronto Face Database; deep neural networks; facial expression; facial landmarks; gradient-based methods; image recognition; learning representations; multiple term objective; multitask learning; Eyebrows; Face; Face recognition; Feature extraction; Neural networks; Standards; Training; computer vision; convolutional neural networks; deep learning; emotion recognition; expression recognition; facial landmarks; multitask learning; representation learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer and Robot Vision (CRV), 2014 Canadian Conference on

Conference_Location

Montreal, QC

Print_ISBN

978-1-4799-4338-8

Type

conf

DOI

10.1109/CRV.2014.21

Filename

6816830