An Attention-based Model for Recognition of Facial Expressions Using CNN-BiLSTM
In recent studies, computer vision and human-computer interaction have focused heavily on facial expression recognition (FER). Traditional deep learning algorithms have faced significant challenges when processing images with occlusion, uneven lighting, and positional inconsistencies and addressing...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Erbil Polytechnic University
2025-02-01
|
| Series: | Polytechnic Journal |
| Subjects: | |
| Online Access: | https://polytechnic-journal.epu.edu.iq/cgi/viewcontent.cgi?article=1849&context=home |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850208696332713984 |
|---|---|
| author | Sushil Kumar Singh Manish Kumar Ikram Majeed Khan A. Jayanthiladevi Chirag Agarwal |
| author_facet | Sushil Kumar Singh Manish Kumar Ikram Majeed Khan A. Jayanthiladevi Chirag Agarwal |
| author_sort | Sushil Kumar Singh |
| collection | DOAJ |
| description | In recent studies, computer vision and human-computer interaction have focused heavily on facial expression
recognition (FER). Traditional deep learning algorithms have faced significant challenges when processing images with
occlusion, uneven lighting, and positional inconsistencies and addressing dataset imbalances. These issues often result
in low accuracy, unreliable recognition rates, and slow convergence. To address the challenges posed by non-frontal
visual features, this study proposes a hybrid model that fusion of Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM), augmented with a point multiplication attention model and Linear
Discriminant Analysis (LDA). Data preparation incorporates median filtering and global contrast normalization to
enhance image quality. The model then utilizes DenseNet and Softmax for image reconstruction, improving feature
maps and providing essential data for classification tasks on the FER2013 and CK þ datasets. We benchmark our proposed model against conventional models such as Convolutional Neural Networks- Long Short-Term Memory (CNNLSTM), Depthwise Separable Convolutional Neural Networks- Long Short-Term Memory (DSCNN-LSTM), Convolutional Neural Network-Bidirectional Long Short-Term Memory (CNN-BiLSTM), and Attention Convolutional Neural
Network - Long Short-Term Memory (ACNN-LSTM), evaluating performance metrics including F1 score, accuracy,
precision, and recall. The results demonstrate that our proposed model outperforms existing approaches, highlighting
the effectiveness of incorporating attention paradigms, hybrid deep learning architectures, and advanced preprocessing
methods for facial emotion detection. The non-parametric statistical test also analyzes it. |
| format | Article |
| id | doaj-art-93be05b9fd334e318db7b0465d4f2f20 |
| institution | OA Journals |
| issn | 2707-7799 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Erbil Polytechnic University |
| record_format | Article |
| series | Polytechnic Journal |
| spelling | doaj-art-93be05b9fd334e318db7b0465d4f2f202025-08-20T02:10:10ZengErbil Polytechnic UniversityPolytechnic Journal2707-77992025-02-011513951https://doi.org/10.59341/2707-7799.1849An Attention-based Model for Recognition of Facial Expressions Using CNN-BiLSTMSushil Kumar Singh0Manish Kumar1Ikram Majeed Khan2A. Jayanthiladevi3Chirag Agarwal4Marwadi University, Rajkot, Gujarat, IndiaSeoul National University of Science and Technology, Seoul, South KoreaCoventry University, Priory St, Coventry, England, UKMarwadi University, Rajkot, Gujarat, IndiaTata Consultancy Services Ltd., USA In recent studies, computer vision and human-computer interaction have focused heavily on facial expression recognition (FER). Traditional deep learning algorithms have faced significant challenges when processing images with occlusion, uneven lighting, and positional inconsistencies and addressing dataset imbalances. These issues often result in low accuracy, unreliable recognition rates, and slow convergence. To address the challenges posed by non-frontal visual features, this study proposes a hybrid model that fusion of Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM), augmented with a point multiplication attention model and Linear Discriminant Analysis (LDA). Data preparation incorporates median filtering and global contrast normalization to enhance image quality. The model then utilizes DenseNet and Softmax for image reconstruction, improving feature maps and providing essential data for classification tasks on the FER2013 and CK þ datasets. We benchmark our proposed model against conventional models such as Convolutional Neural Networks- Long Short-Term Memory (CNNLSTM), Depthwise Separable Convolutional Neural Networks- Long Short-Term Memory (DSCNN-LSTM), Convolutional Neural Network-Bidirectional Long Short-Term Memory (CNN-BiLSTM), and Attention Convolutional Neural Network - Long Short-Term Memory (ACNN-LSTM), evaluating performance metrics including F1 score, accuracy, precision, and recall. The results demonstrate that our proposed model outperforms existing approaches, highlighting the effectiveness of incorporating attention paradigms, hybrid deep learning architectures, and advanced preprocessing methods for facial emotion detection. The non-parametric statistical test also analyzes it.https://polytechnic-journal.epu.edu.iq/cgi/viewcontent.cgi?article=1849&context=homefacial expression recognition,cnn,bilstm,attention model,data preprocessing,global contrast normalization |
| spellingShingle | Sushil Kumar Singh Manish Kumar Ikram Majeed Khan A. Jayanthiladevi Chirag Agarwal An Attention-based Model for Recognition of Facial Expressions Using CNN-BiLSTM Polytechnic Journal facial expression recognition, cnn, bilstm, attention model, data preprocessing, global contrast normalization |
| title | An Attention-based Model for Recognition of Facial Expressions Using CNN-BiLSTM |
| title_full | An Attention-based Model for Recognition of Facial Expressions Using CNN-BiLSTM |
| title_fullStr | An Attention-based Model for Recognition of Facial Expressions Using CNN-BiLSTM |
| title_full_unstemmed | An Attention-based Model for Recognition of Facial Expressions Using CNN-BiLSTM |
| title_short | An Attention-based Model for Recognition of Facial Expressions Using CNN-BiLSTM |
| title_sort | attention based model for recognition of facial expressions using cnn bilstm |
| topic | facial expression recognition, cnn, bilstm, attention model, data preprocessing, global contrast normalization |
| url | https://polytechnic-journal.epu.edu.iq/cgi/viewcontent.cgi?article=1849&context=home |
| work_keys_str_mv | AT sushilkumarsingh anattentionbasedmodelforrecognitionoffacialexpressionsusingcnnbilstm AT manishkumar anattentionbasedmodelforrecognitionoffacialexpressionsusingcnnbilstm AT ikrammajeedkhan anattentionbasedmodelforrecognitionoffacialexpressionsusingcnnbilstm AT ajayanthiladevi anattentionbasedmodelforrecognitionoffacialexpressionsusingcnnbilstm AT chiragagarwal anattentionbasedmodelforrecognitionoffacialexpressionsusingcnnbilstm AT sushilkumarsingh attentionbasedmodelforrecognitionoffacialexpressionsusingcnnbilstm AT manishkumar attentionbasedmodelforrecognitionoffacialexpressionsusingcnnbilstm AT ikrammajeedkhan attentionbasedmodelforrecognitionoffacialexpressionsusingcnnbilstm AT ajayanthiladevi attentionbasedmodelforrecognitionoffacialexpressionsusingcnnbilstm AT chiragagarwal attentionbasedmodelforrecognitionoffacialexpressionsusingcnnbilstm |