Next Article in Journal
Optimal Reconstruction of Single-Pixel Images through Feature Feedback Mechanism and Attention
Next Article in Special Issue
Face Detection Based on DF-Net
Previous Article in Journal
Sound Event Detection with Perturbed Residual Recurrent Neural Network
Previous Article in Special Issue
Locality-Sensitive Hashing of Soft Biometrics for Efficient Face Image Database Search and Retrieval
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Facial Expression Recognition in the Wild for Low-Resolution Images Using Voting Residual Network

by
José L. Gómez-Sirvent
1,2,
Francisco López de la Rosa
1,3,
María T. López
1,2 and
Antonio Fernández-Caballero
1,2,4,*
1
Instituto de Investigación en Informática de Albacete, Universidad de Castilla-La Mancha, 02071 Albacete, Spain
2
Departamento de Sistemas Informáticos, Universidad de Castilla-La Mancha, 02071 Albacete, Spain
3
Departamento de Ingeniería Eléctrica, Electrónica, Automática y Comunicaciones, Universidad de Castilla-La Mancha, 02071 Albacete, Spain
4
CIBERSAM-ISCIII (Biomedical Research Networking Center in Mental Health, Instituto de Salud Carlos III), 28016 Madrid, Spain
*
Author to whom correspondence should be addressed.
Electronics 2023, 12(18), 3837; https://doi.org/10.3390/electronics12183837
Submission received: 25 July 2023 / Revised: 6 September 2023 / Accepted: 8 September 2023 / Published: 11 September 2023
(This article belongs to the Special Issue Intelligent Face Recognition and Multiple Applications)

Abstract

Facial expression recognition (FER) in the wild has attracted much attention in recent years due to its wide range of applications. Most current approaches use deep learning models trained on relatively large images, which significantly reduces their accuracy when they have to infer low-resolution images. In this paper, a residual voting network is proposed for the classification of low-resolution facial expression images. Specifically, the network consists of a modified ResNet-18, which divides each sample into multiple overlapping crops, makes a prediction of the class to which each of the crops belongs, and by soft-voting the predictions of all the crops, the network determines the class of the sample. A novel aspect of this work is that the image splitting is not performed before entering the network, but at an intermediate point in the network, which significantly reduces the resource consumption. The proposed approach was evaluated on two popular benchmark datasets (AffectNet and RAF-DB) by scaling the images to a network input size of 48 × 48. The proposed model reported an accuracy of 63.06% on AffectNet and 85.69% on RAF-DB with seven classes in both cases, which are values comparable to those provided by other current approaches using much larger images.
Keywords: facial expression recognition; emotions; low resolution; AffectNet; RAF-DB; voting facial expression recognition; emotions; low resolution; AffectNet; RAF-DB; voting

Share and Cite

MDPI and ACS Style

Gómez-Sirvent, J.L.; López de la Rosa, F.; López, M.T.; Fernández-Caballero, A. Facial Expression Recognition in the Wild for Low-Resolution Images Using Voting Residual Network. Electronics 2023, 12, 3837. https://doi.org/10.3390/electronics12183837

AMA Style

Gómez-Sirvent JL, López de la Rosa F, López MT, Fernández-Caballero A. Facial Expression Recognition in the Wild for Low-Resolution Images Using Voting Residual Network. Electronics. 2023; 12(18):3837. https://doi.org/10.3390/electronics12183837

Chicago/Turabian Style

Gómez-Sirvent, José L., Francisco López de la Rosa, María T. López, and Antonio Fernández-Caballero. 2023. "Facial Expression Recognition in the Wild for Low-Resolution Images Using Voting Residual Network" Electronics 12, no. 18: 3837. https://doi.org/10.3390/electronics12183837

APA Style

Gómez-Sirvent, J. L., López de la Rosa, F., López, M. T., & Fernández-Caballero, A. (2023). Facial Expression Recognition in the Wild for Low-Resolution Images Using Voting Residual Network. Electronics, 12(18), 3837. https://doi.org/10.3390/electronics12183837

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop