TY - GEN
T1 - On the Use of Convolutional Neural Network Architectures for Facial Emotion Recognition
AU - Espinel, Andrés
AU - Pérez, Noel
AU - Riofrío, Daniel
AU - Benítez, Diego S.
AU - Moyano, Ricardo Flores
N1 - Publisher Copyright:
© 2022, Springer Nature Switzerland AG.
PY - 2022
Y1 - 2022
N2 - This work compares face gesture recognition methods based on deep learning convolutional neural network and autoencoder architectures named DCNN1, DCNN2, DCNN3, DCNN4, and DCNN+Autoencoder, that maximize the classification performance on single and mixing databases. We validated the proposed architectures on four different databases: Jaffe, CK+, FACES, and the combination of them over a five-fold cross-validation strategy. The DCNN4 was the best model in the Jaffe and FACES databases, obtaining accuracy scores of 95 % and 97 %, respectively. The DCNN2 achieved the best accuracy performance of 94 % in the CK+ database. Finally, the DCNN+Autoencoder stands as the best model in the combination of all databases (Jaffe & CK+ & FACES), achieving an accuracy score of 92 %. Moreover, according to the cross-entropy loss function, the best model per database did not incur overfitting.
AB - This work compares face gesture recognition methods based on deep learning convolutional neural network and autoencoder architectures named DCNN1, DCNN2, DCNN3, DCNN4, and DCNN+Autoencoder, that maximize the classification performance on single and mixing databases. We validated the proposed architectures on four different databases: Jaffe, CK+, FACES, and the combination of them over a five-fold cross-validation strategy. The DCNN4 was the best model in the Jaffe and FACES databases, obtaining accuracy scores of 95 % and 97 %, respectively. The DCNN2 achieved the best accuracy performance of 94 % in the CK+ database. Finally, the DCNN+Autoencoder stands as the best model in the combination of all databases (Jaffe & CK+ & FACES), achieving an accuracy score of 92 %. Moreover, according to the cross-entropy loss function, the best model per database did not incur overfitting.
KW - Artificial intelligence
KW - Deep-learning models
KW - Face emotion recognition
KW - Face gesture classification
KW - Face images
UR - http://www.scopus.com/inward/record.url?scp=85126259555&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-91308-3_2
DO - 10.1007/978-3-030-91308-3_2
M3 - Contribución a la conferencia
AN - SCOPUS:85126259555
SN - 9783030913076
T3 - Communications in Computer and Information Science
SP - 18
EP - 30
BT - Applications of Computational Intelligence - 4th IEEE Colombian Conference, ColCACI 2021, Revised Selected Papers
A2 - Orjuela-Cañón, Alvaro David
A2 - Lopez, Jesus A.
A2 - Arias-Londoño, Julián David
A2 - Figueroa-García, Juan Carlos
PB - Springer Science and Business Media Deutschland GmbH
T2 - 4th IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI 2021
Y2 - 27 May 2021 through 28 May 2021
ER -