Ad-Corre: Adaptive Correlation-Based Loss for Facial Expression Recognition in the Wild
Automated Facial Expression Recognition (FER) in the wild using deep neural networks is still challenging due to intra-class variations and inter-class similarities in facial images. Deep Metric Learning (DML) is among the widely used methods to deal with these issues by improving the discriminative...
Main Authors: | , |
---|---|
Format: | Text |
Language: | unknown |
Published: |
Digital Commons @ DU
2022
|
Subjects: | |
Online Access: | https://digitalcommons.du.edu/electrical_engineering_faculty/53 https://digitalcommons.du.edu/cgi/viewcontent.cgi?article=1052&context=electrical_engineering_faculty |
id |
ftunivdenverir:oai:digitalcommons.du.edu:electrical_engineering_faculty-1052 |
---|---|
record_format |
openpolar |
spelling |
ftunivdenverir:oai:digitalcommons.du.edu:electrical_engineering_faculty-1052 2023-06-06T11:53:10+02:00 Ad-Corre: Adaptive Correlation-Based Loss for Facial Expression Recognition in the Wild Fard, Ali Pourramezan Mahoor, Mohammad H 2022-03-03T08:00:00Z application/pdf https://digitalcommons.du.edu/electrical_engineering_faculty/53 https://digitalcommons.du.edu/cgi/viewcontent.cgi?article=1052&context=electrical_engineering_faculty unknown Digital Commons @ DU https://digitalcommons.du.edu/electrical_engineering_faculty/53 https://digitalcommons.du.edu/cgi/viewcontent.cgi?article=1052&context=electrical_engineering_faculty http://creativecommons.org/licenses/by-nc-nd/4.0/ Electrical and Computer Engineering: Faculty Scholarship Facial expression recognition Facial emotion recognition Ad-Corre loss Loss function Convolutional neural network Daniel Felix Ritchie School of Engineering and Computer Science Electrical and Computer Engineering Engineering OS and Networks Other Electrical and Computer Engineering Software Engineering text 2022 ftunivdenverir 2023-04-13T17:54:31Z Automated Facial Expression Recognition (FER) in the wild using deep neural networks is still challenging due to intra-class variations and inter-class similarities in facial images. Deep Metric Learning (DML) is among the widely used methods to deal with these issues by improving the discriminative power of the learned embedded features. This paper proposes an Adaptive Correlation (Ad-Corre) Loss to guide the network towards generating embedded feature vectors with high correlation for within-class samples and less correlation for between-class samples. Ad-Corre consists of 3 components called Feature Discriminator, Mean Discriminator, and Embedding Discriminator. We design the Feature Discriminator component to guide the network to create the embedded feature vectors to be highly correlated if they belong to a similar class, and less correlated if they belong to different classes. In addition, the Mean Discriminator component leads the network to make the mean embedded feature vectors of different classes to be less similar to each other. We use Xception network as the backbone of our model, and contrary to previous work, we propose an embedding feature space that contains k feature vectors. Then, the Embedding Discriminator component penalizes the network to generate the embedded feature vectors, which are dissimilar. We trained our model using the combination of our proposed loss functions called Ad-Corre Loss jointly with the crossentropy loss. We achieved a very promising recognition accuracy on AffectNet, RAF-DB, and FER-2013. Our extensive experiments and ablation study indicate the power of our method to cope well with challenging FER tasks in the wild. The code is available on Github. Text DML University of Denver: Digital Commons @ DU Ritchie ENVELOPE(158.417,158.417,-78.533,-78.533) |
institution |
Open Polar |
collection |
University of Denver: Digital Commons @ DU |
op_collection_id |
ftunivdenverir |
language |
unknown |
topic |
Facial expression recognition Facial emotion recognition Ad-Corre loss Loss function Convolutional neural network Daniel Felix Ritchie School of Engineering and Computer Science Electrical and Computer Engineering Engineering OS and Networks Other Electrical and Computer Engineering Software Engineering |
spellingShingle |
Facial expression recognition Facial emotion recognition Ad-Corre loss Loss function Convolutional neural network Daniel Felix Ritchie School of Engineering and Computer Science Electrical and Computer Engineering Engineering OS and Networks Other Electrical and Computer Engineering Software Engineering Fard, Ali Pourramezan Mahoor, Mohammad H Ad-Corre: Adaptive Correlation-Based Loss for Facial Expression Recognition in the Wild |
topic_facet |
Facial expression recognition Facial emotion recognition Ad-Corre loss Loss function Convolutional neural network Daniel Felix Ritchie School of Engineering and Computer Science Electrical and Computer Engineering Engineering OS and Networks Other Electrical and Computer Engineering Software Engineering |
description |
Automated Facial Expression Recognition (FER) in the wild using deep neural networks is still challenging due to intra-class variations and inter-class similarities in facial images. Deep Metric Learning (DML) is among the widely used methods to deal with these issues by improving the discriminative power of the learned embedded features. This paper proposes an Adaptive Correlation (Ad-Corre) Loss to guide the network towards generating embedded feature vectors with high correlation for within-class samples and less correlation for between-class samples. Ad-Corre consists of 3 components called Feature Discriminator, Mean Discriminator, and Embedding Discriminator. We design the Feature Discriminator component to guide the network to create the embedded feature vectors to be highly correlated if they belong to a similar class, and less correlated if they belong to different classes. In addition, the Mean Discriminator component leads the network to make the mean embedded feature vectors of different classes to be less similar to each other. We use Xception network as the backbone of our model, and contrary to previous work, we propose an embedding feature space that contains k feature vectors. Then, the Embedding Discriminator component penalizes the network to generate the embedded feature vectors, which are dissimilar. We trained our model using the combination of our proposed loss functions called Ad-Corre Loss jointly with the crossentropy loss. We achieved a very promising recognition accuracy on AffectNet, RAF-DB, and FER-2013. Our extensive experiments and ablation study indicate the power of our method to cope well with challenging FER tasks in the wild. The code is available on Github. |
format |
Text |
author |
Fard, Ali Pourramezan Mahoor, Mohammad H |
author_facet |
Fard, Ali Pourramezan Mahoor, Mohammad H |
author_sort |
Fard, Ali Pourramezan |
title |
Ad-Corre: Adaptive Correlation-Based Loss for Facial Expression Recognition in the Wild |
title_short |
Ad-Corre: Adaptive Correlation-Based Loss for Facial Expression Recognition in the Wild |
title_full |
Ad-Corre: Adaptive Correlation-Based Loss for Facial Expression Recognition in the Wild |
title_fullStr |
Ad-Corre: Adaptive Correlation-Based Loss for Facial Expression Recognition in the Wild |
title_full_unstemmed |
Ad-Corre: Adaptive Correlation-Based Loss for Facial Expression Recognition in the Wild |
title_sort |
ad-corre: adaptive correlation-based loss for facial expression recognition in the wild |
publisher |
Digital Commons @ DU |
publishDate |
2022 |
url |
https://digitalcommons.du.edu/electrical_engineering_faculty/53 https://digitalcommons.du.edu/cgi/viewcontent.cgi?article=1052&context=electrical_engineering_faculty |
long_lat |
ENVELOPE(158.417,158.417,-78.533,-78.533) |
geographic |
Ritchie |
geographic_facet |
Ritchie |
genre |
DML |
genre_facet |
DML |
op_source |
Electrical and Computer Engineering: Faculty Scholarship |
op_relation |
https://digitalcommons.du.edu/electrical_engineering_faculty/53 https://digitalcommons.du.edu/cgi/viewcontent.cgi?article=1052&context=electrical_engineering_faculty |
op_rights |
http://creativecommons.org/licenses/by-nc-nd/4.0/ |
_version_ |
1767959286230548480 |