EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Towards a Better Performance in Facial Expression Recognition: A Data‐Centric Approach

Christian Mejia-Escobar iD, Miguel Cazorla iD, Ester Martinez-Martin iD

Computational Intelligence and Neuroscience · 2023

DOI: 10.1155/2023/1394882 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computational Intelligence and Neuroscience, 2023. Verfasst von Christian Mejia-Escobar und 2 weiteren Autor:innen. Beteiligte Einrichtung: Central University of Ecuador, P.O. Box 17-03-100, Quito, Ecuador. Hauptthemen laut MEDLINE: Facial Recognition, Robotics. Die Arbeit wurde bisher 17-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Conselleria de Innovación, Universidades, Ciencia y Sociedad Digital, Generalitat Valenciana, Universidad Central del Ecuador.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Facial expression is the best evidence of our emotions. Its automatic detection and recognition are key for robotics, medicine, healthcare, education, psychology, sociology, marketing, security, entertainment, and many other areas. Experiments in the lab environments achieve high performance. However, in real‐world scenarios, it is challenging. Deep learning techniques based on convolutional neural networks (CNNs) have shown great potential. Most of the research is exclusively model‐centric, searching for better algorithms to improve recognition. However, progress is insufficient. Despite being the main resource for automatic learning, few works focus on improving the quality of datasets. We propose a novel data‐centric method to tackle misclassification, a problem commonly encountered in facial image datasets. The strategy is to progressively refine the dataset by successive training of a CNN model that is fixed. Each training uses the facial images corresponding to the correct predictions of the previous training, allowing the model to capture more distinctive features of each class of facial expression. After the last training, the model performs automatic reclassification of the whole dataset. Unlike other similar work, our method avoids modifying, deleting, or augmenting facial images. Experimental results on three representative datasets proved the effectiveness of the proposed method, improving the validation accuracy by 20.45%, 14.47%, and 39.66%, for FER2013, NHFI, and AffectNet, respectively. The recognition rates on the reclassified versions of these datasets are 86.71%, 70.44%, and 89.17% and become state‐of‐the‐art performance.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Facial RecognitionRoboticsNeural Networks, ComputerAlgorithmsFaceFacial Expression

Autor:innen und Institutionen

  • Central University of Ecuador, P.O. Box 17-03-100, Quito, Ecuador.
  • Institute for Computer Research, University of Alicante, P.O. Box 99. 03080, Alicante, Spain.

ORCID-Profile: Christian Mejia-Escobar, Miguel Cazorla, Ester Martinez-Martin

Förderung

  • Conselleria de Innovación, Universidades, Ciencia y Sociedad Digital, Generalitat Valenciana
  • Universidad Central del Ecuador

Zitieren

APA

Mejia-Escobar, C., Cazorla, M., & Martinez-Martin, E. (2023). Towards a Better Performance in Facial Expression Recognition: A Data‐Centric Approach. Computational Intelligence and Neuroscience. https://doi.org/10.1155/2023/1394882

Vancouver

Mejia-Escobar C, Cazorla M, Martinez-Martin E. Towards a Better Performance in Facial Expression Recognition: A Data‐Centric Approach. Computational Intelligence and Neuroscience. 2023. doi:10.1155/2023/1394882

BibTeX
@article{MejiaEscobar2023,
  author = {Mejia-Escobar, Christian and Cazorla, Miguel and Martinez-Martin, Ester},
  title = {Towards a Better Performance in Facial Expression Recognition: A Data‐Centric Approach},
  journal = {Computational Intelligence and Neuroscience},
  year = {2023},
  doi = {10.1155/2023/1394882}
}
Kostenlos starten →