eCite Digital Repository

Facial emotion recognition using an ensemble of multi-level convolutional neural networks

Citation

Nguyen, HD and Yeom, S and Lee, G-S and Yang, H-J and Na, I-S and Kim, S-H, Facial emotion recognition using an ensemble of multi-level convolutional neural networks, International Journal of Pattern Recognition and Artificial Intelligence pp. 1-17. ISSN 0218-0014 (2019) [Refereed Article]

Copyright Statement

Copyright 2019 World Scientific Publishing Company

DOI: doi:10.1142/S0218001419400159

Abstract

Emotion recognition plays an indispensable role in human-machine interaction system. The process includes finding interesting facial regions in images and classifying them into one of seven classes: angry, disgust, fear, happy, neutral, sad, and surprise. Although many breakthroughs have been made in image classification, especially in facial expression recognition, this research area is still challenging in terms of wild sampling environment. In this paper, we used multi-level features in a convolutional neural network for facial expression recognition. Based on our observations, we introduced various network connections to improve the classification task. By combining the proposed network connections, our method achieved competitive results compared to state-of-the-art methods on the FER2013 dataset.

Item Details

Item Type:Refereed Article
Keywords:ensemble model, facial emotion recognition in the wild, multi-level convolutional neural networks
Research Division:Information and Computing Sciences
Research Group:Artificial Intelligence and Image Processing
Research Field:Pattern Recognition and Data Mining
Objective Division:Information and Communication Services
Objective Group:Computer Software and Services
Objective Field:Information Processing Services (incl. Data Entry and Capture)
UTAS Author:Yeom, S (Dr Soonja Yeom)
ID Code:132151
Year Published:2019
Deposited By:Information and Communication Technology
Deposited On:2019-04-24
Last Modified:2019-06-20
Downloads:0

Repository Staff Only: item control page