| Survey on Emotion Recognition Using Deep Learning |
| Paper ID : 1042-ICEEM2025 (R2) |
| Authors |
|
shaimaa Elsayed Hassan *1, somaya feshawy2, Moawad Dessouky2, Hamdy Shrshr2 1faculty of electronic engineering
Electronic and Electrical Communication Department 2faculty of electronic engineering |
| Abstract |
| Facial Emotion Recognition (FER), a vital branch of affective computing, enables machines to interpret human emotions through facial expressions. Given that facial cues account for approximately 55% of the emotional content in face-to-face communication, FER serves as a critical bridge between human and machine interaction. Advances in this field are particularly important for developing responsive robotic and computer systems. Emotion recognition using deep learning has become an area of increasing interest in recent years, owing to its potential applications in sectors such as human-computer interaction, healthcare, and security. This review offers a thorough overview of the current methods, approaches, and challenges in emotion recognition through deep learning. It explores various tasks related to emotion recognition, such as facial expression analysis, speech emotion detection, and processing physiological signals. The paper provides an in-depth analysis of the deep learning models typically employed in emotion recognition, including Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Transformer-based models. It also emphasizes the role of large, annotated datasets, cross-domain generalization, and multimodal integration in enhancing the accuracy and robustness of emotion recognition systems. Furthermore, the survey addresses ethical issues, biases, and privacy concerns surrounding emotion recognition technologies. Lastly, the paper points to future research opportunities and advancements necessary to improve the performance and broader applicability of deep learning-based emotion recognition systems |
| Keywords |
| Artificial Intelligence, Emotion Detection, Machine Learning, Facial Recognition, Datasets. |
| Status: Accepted |