Emotion recognition in the wild with feature fusion and multiple Kernel learning

Junkai Chen; Zenghai Chen; Zheru Chi; Hong Fu

doi:10.1145/2663204.2666277

Emotion recognition in the wild with feature fusion and multiple Kernel learning

Junkai Chen
, Zenghai Chen
, Zheru Chi
, Hong Fu

The Hong Kong Polytechnic University

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

89 Citations (Scopus)

Abstract

2014 ACM. This paper presents our proposed approach for the second Emotion Recognition in The Wild Challenge. We propose a new feature descriptor called Histogram of Oriented Gradients from Three Orthogonal Planes (HOG-TOP) to represent facial expressions. We also explore the properties of visual features and audio features, and adopt Multiple Kernel Learning (MKL) to find an optimal feature fusion. An SVM with multiple kernels is trained for the facial expression classification. Experimental results demonstrate that our method achieves a promising performance. The overall classification accuracy on the validation set and test set are 40.21% and 45.21%, respectively.

Original language	English
Title of host publication	ICMI 2014 - Proceedings of the 2014 International Conference on Multimodal Interaction
Publisher	Association for Computing Machinery, Inc
Pages	508-513
Number of pages	6
ISBN (Electronic)	9781450328852
DOIs	https://doi.org/10.1145/2663204.2666277
Publication status	Published - 12 Nov 2014
Event	16th ACM International Conference on Multimodal Interaction, ICMI 2014 - Istanbul, Turkey Duration: 12 Nov 2014 → 16 Nov 2014

Conference

Conference	16th ACM International Conference on Multimodal Interaction, ICMI 2014
Country/Territory	Turkey
City	Istanbul
Period	12/11/14 → 16/11/14

Keywords

Emotion recognition
Feature fusion
HOG-TOP
Multiple kernel learning
Support vector machine

ASJC Scopus subject areas

Human-Computer Interaction

More information

10.1145/2663204.2666277

Cite this

@inproceedings{d4ff15463bb74355a1733b74c8954823,

title = "Emotion recognition in the wild with feature fusion and multiple Kernel learning",

abstract = "2014 ACM. This paper presents our proposed approach for the second Emotion Recognition in The Wild Challenge. We propose a new feature descriptor called Histogram of Oriented Gradients from Three Orthogonal Planes (HOG-TOP) to represent facial expressions. We also explore the properties of visual features and audio features, and adopt Multiple Kernel Learning (MKL) to find an optimal feature fusion. An SVM with multiple kernels is trained for the facial expression classification. Experimental results demonstrate that our method achieves a promising performance. The overall classification accuracy on the validation set and test set are 40.21\% and 45.21\%, respectively.",

keywords = "Emotion recognition, Feature fusion, HOG-TOP, Multiple kernel learning, Support vector machine",

author = "Junkai Chen and Zenghai Chen and Zheru Chi and Hong Fu",

year = "2014",

month = nov,

day = "12",

doi = "10.1145/2663204.2666277",

language = "English",

pages = "508--513",

booktitle = "ICMI 2014 - Proceedings of the 2014 International Conference on Multimodal Interaction",

publisher = "Association for Computing Machinery, Inc",

note = "16th ACM International Conference on Multimodal Interaction, ICMI 2014 ; Conference date: 12-11-2014 Through 16-11-2014",

}

Chen, J, Chen, Z, Chi, Z & Fu, H 2014, Emotion recognition in the wild with feature fusion and multiple Kernel learning. in ICMI 2014 - Proceedings of the 2014 International Conference on Multimodal Interaction. Association for Computing Machinery, Inc, pp. 508-513, 16th ACM International Conference on Multimodal Interaction, ICMI 2014, Istanbul, Turkey, 12/11/14. https://doi.org/10.1145/2663204.2666277

Emotion recognition in the wild with feature fusion and multiple Kernel learning. / Chen, Junkai; Chen, Zenghai; Chi, Zheru et al.
ICMI 2014 - Proceedings of the 2014 International Conference on Multimodal Interaction. Association for Computing Machinery, Inc, 2014. p. 508-513.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Emotion recognition in the wild with feature fusion and multiple Kernel learning

AU - Chen, Junkai

AU - Chen, Zenghai

AU - Chi, Zheru

AU - Fu, Hong

PY - 2014/11/12

Y1 - 2014/11/12

N2 - 2014 ACM. This paper presents our proposed approach for the second Emotion Recognition in The Wild Challenge. We propose a new feature descriptor called Histogram of Oriented Gradients from Three Orthogonal Planes (HOG-TOP) to represent facial expressions. We also explore the properties of visual features and audio features, and adopt Multiple Kernel Learning (MKL) to find an optimal feature fusion. An SVM with multiple kernels is trained for the facial expression classification. Experimental results demonstrate that our method achieves a promising performance. The overall classification accuracy on the validation set and test set are 40.21% and 45.21%, respectively.

AB - 2014 ACM. This paper presents our proposed approach for the second Emotion Recognition in The Wild Challenge. We propose a new feature descriptor called Histogram of Oriented Gradients from Three Orthogonal Planes (HOG-TOP) to represent facial expressions. We also explore the properties of visual features and audio features, and adopt Multiple Kernel Learning (MKL) to find an optimal feature fusion. An SVM with multiple kernels is trained for the facial expression classification. Experimental results demonstrate that our method achieves a promising performance. The overall classification accuracy on the validation set and test set are 40.21% and 45.21%, respectively.

KW - Emotion recognition

KW - Feature fusion

KW - HOG-TOP

KW - Multiple kernel learning

KW - Support vector machine

UR - https://www.scopus.com/pages/publications/84947274167

U2 - 10.1145/2663204.2666277

DO - 10.1145/2663204.2666277

M3 - Conference article published in proceeding or book

SP - 508

EP - 513

BT - ICMI 2014 - Proceedings of the 2014 International Conference on Multimodal Interaction

PB - Association for Computing Machinery, Inc

T2 - 16th ACM International Conference on Multimodal Interaction, ICMI 2014

Y2 - 12 November 2014 through 16 November 2014

ER -

Emotion recognition in the wild with feature fusion and multiple Kernel learning

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this