Arnab Kumar Roy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Arnab Kumar Roy has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Abhishek Dey
Bay Area Advanced Analytics India (P) Ltd., A Kaliber.AI company, Guwahati, India
1 shared publication Adhitiya Sharma
National Institute of Technology Sikkim, Ravangla, India
1 shared publication- Hemant Kumar Kathania
National Institute of Technology Sikkim, Ravangla, India
1 shared publication - Sarfaraj Alam Ansari
Department of Computer Science and Engineering, National Institute of Technology Sikkim, Ravangla, India
1 shared publication
Latest Publications
- 23 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 32 (10709908) , 491-495
ResEmoteNet: Bridging Accuracy and Loss Reduction in Facial Emotion Recognition
The human face is a silent communicator, expressing emotions and thoughts through it's facial expressions. With the advancements in computer vision in recent years, facial emotion recognition technology has made significant strides, enabling machines to decode the intricacies of facial cues. In this work, we propose ResEmoteNet, a novel deep learning architecture for facial emotion recognition designed with the combination of Convolutional, Squeeze-Excitation (SE) and Residual Networks. The inclusion of SE block selectively focuses on the important features of the human face, enhances the feature representation and suppresses the less relevant ones. This helps in reducing the loss and enhancing the overall model performance. We also integrate the SE block with three residual blocks that help in learning more complex representation of the data through deeper layers. We evaluated ResEmoteNet on four open-source databases: FER2013, RAF-DB, AffectNet-7 and ExpW, achieving accuracies of 79.79%, 94.76%, 72.39% and 75.67% respectively. The proposed network outperforms state-of-the-art models across all four databases.
- 23 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 32 (10709908) , 491-495
ResEmoteNet: Bridging Accuracy and Loss Reduction in Facial Emotion Recognition
The human face is a silent communicator, expressing emotions and thoughts through it's facial expressions. With the advancements in computer vision in recent years, facial emotion recognition technology has made significant strides, enabling machines to decode the intricacies of facial cues. In this work, we propose ResEmoteNet, a novel deep learning architecture for facial emotion recognition designed with the combination of Convolutional, Squeeze-Excitation (SE) and Residual Networks. The inclusion of SE block selectively focuses on the important features of the human face, enhances the feature representation and suppresses the less relevant ones. This helps in reducing the loss and enhancing the overall model performance. We also integrate the SE block with three residual blocks that help in learning more complex representation of the data through deeper layers. We evaluated ResEmoteNet on four open-source databases: FER2013, RAF-DB, AffectNet-7 and ExpW, achieving accuracies of 79.79%, 94.76%, 72.39% and 75.67% respectively. The proposed network outperforms state-of-the-art models across all four databases.