EmoNets: Multimodal deep learning approaches for emotion recognition in video

The task of the Emotion Recognition in the Wild (EmotiW) Challenge is to assign one of seven emotions to short video clips extracted from Hollywood style movies. The videos depict acted-out emotions under realistic conditions with a large degree of variation in attributes such as pose and illuminati...

Full description

Saved in:
Bibliographic Details
Published in:Journal on multimodal user interfaces 2016-06, Vol.10 (2), p.99-111
Main Authors: Kahou, Samira Ebrahimi, Bouthillier, Xavier, Lamblin, Pascal, Gulcehre, Caglar, Michalski, Vincent, Konda, Kishore, Jean, Sébastien, Froumenty, Pierre, Dauphin, Yann, Boulanger-Lewandowski, Nicolas, Chandias Ferrari, Raul, Mirza, Mehdi, Warde-Farley, David, Courville, Aaron, Vincent, Pascal, Memisevic, Roland, Pal, Christopher, Bengio, Yoshua
Format: Article
Language:English
Subjects:
Computer Science
Image Processing and Computer Vision
Original Paper
Signal,Image and Speech Processing
User Interfaces and Human Computer Interaction
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The task of the Emotion Recognition in the Wild (EmotiW) Challenge is to assign one of seven emotions to short video clips extracted from Hollywood style movies. The videos depict acted-out emotions under realistic conditions with a large degree of variation in attributes such as pose and illumination, making it worthwhile to explore approaches which consider combinations of features from multiple modalities for label assignment. In this paper we present our approach to learning several specialist models using deep learning techniques, each focusing on one modality. Among these are a convolutional neural network, focusing on capturing visual information in detected faces, a deep belief net focusing on the representation of the audio stream, a K-Means based “bag-of-mouths” model, which extracts visual features around the mouth region and a relational autoencoder, which addresses spatio-temporal aspects of videos. We explore multiple methods for the combination of cues from these modalities into one common classifier. This achieves a considerably greater accuracy than predictions from our strongest single-modality classifier. Our method was the winning submission in the 2013 EmotiW challenge and achieved a test set accuracy of 47.67 % on the 2014 dataset.
ISSN:1783-7677
1783-8738
DOI:10.1007/s12193-015-0195-2