Paper
4 January 2021 The impact of pre-processing algorithms in facial expression recognition
Daniel Canedo, António J. R. Neves
Author Affiliations +
Proceedings Volume 11605, Thirteenth International Conference on Machine Vision; 116051Q (2021) https://doi.org/10.1117/12.2587865
Event: Thirteenth International Conference on Machine Vision, 2020, Rome, Italy
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.
© (2021) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Daniel Canedo and António J. R. Neves "The impact of pre-processing algorithms in facial expression recognition", Proc. SPIE 11605, Thirteenth International Conference on Machine Vision, 116051Q (4 January 2021); https://doi.org/10.1117/12.2587865
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
Back to Top