Please use this identifier to cite or link to this item:
Title: The impact of pre-processing algorithms in facial expression recognition
Author: Canedo, Daniel
Neves, António J. R.
Keywords: Deep learning
Facial expression recognition
Computer vision
Attention maps
Convolutional neural networks
Issue Date: 2020
Publisher: Society of Photo-optical Instrumentation Engineers
Abstract: This paper proposes several pre-processing algorithms to improve facial expression recognition based on Convolutional Neural Networks (CNNs) models. The proposed CNN model was trained on the Extended Cohn-Kanade dataset (CK+) after applying the pre-processing stages and achieved competitive results (93.90% recognition accuracy) despite its simple and light architecture. Using this CNN model, a study on the impact of each pre-processing algorithm when extracting facial features is presented. In the end, it is understood that pre-processing algorithms help CNNs to extract the most relevant features for each facial expression more effectively, reducing the overfitting and increasing the recognition accuracy. Attention maps before and after the pre-processing step are shown in order to visualize its impact when the proposed CNN model makes a prediction.
Peer review: yes
DOI: 10.1117/12.2587865
ISSN: 0277-786X
Appears in Collections:DETI - Artigos
IEETA - Artigos

Files in This Item:
File Description SizeFormat 
ICMV 2020_final_draft.pdf2.08 MBAdobe PDFrestrictedAccess

Formato BibTex MendeleyEndnote Degois 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.