Deep Learning for Emotion Recognition in Faces

Ariel Ruiz-Garcia, Mark Elshaw, Abdulrahman Altahhan, Vasile Palade

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

16 Citations (Scopus)


Deep Learning (DL) has shown real promise for the classification efficiency for emotion recognition problems. In this paper we present experimental results for a deeply-trained model for emotion recognition through the use of facial expression images. We explore two Convolutional Neural Network (CNN) architectures that offer automatic feature extraction and representation, followed by fully connected softmax layers to classify images into seven emotions. The first architecture explores the impact of reducing the number of deep learning layers and the second splits the input images horizontally into two streams based on eye and mouth positions. The first proposed architecture produces state of the art results with an accuracy rate of 96.93 % and the second architecture with split input produces an average accuracy rate of 86.73 %, respectively.
Original languageEnglish
Title of host publicationArtificial Neural Networks and Machine Learning – ICANN 2016
EditorsAlessandro E.P. Villa, Paolo Masulli, Antonio Javier Pons Rivero
Place of PublicationSwitzerland
PublisherSpringer Verlag
ISBN (Print)978-3-319-44780-3, 978-3-319-44781-0
Publication statusPublished - 13 Aug 2016
EventThe 25th International Conference on Artificial Neural Networks - Barcelona, Spain
Duration: 6 Sep 20169 Sep 2016


ConferenceThe 25th International Conference on Artificial Neural Networks
Abbreviated titleICANN 2016

Bibliographical note

The full text is not available on the repository.


  • Deep learning
  • Convolution neural networks
  • Emotion recognition
  • Empathic robots


Dive into the research topics of 'Deep Learning for Emotion Recognition in Faces'. Together they form a unique fingerprint.

Cite this