To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Detecting Emotions from Facial Features Using Deep Learning and Computer Vision |
|
Author |
Amal Alshahrani, Shahad Abdullah Alzahrani, Atheer Talal AL-Sharief, Najat Ali Alshehri, Majd Hasan Alhakami, Shahd Amein Wayyani |
| Citation |
Vol. 25 No. 5 pp. 79-90 |
|
Abstract |
Emotion detection from facial expressions is a crucial task with applications in psychology, human-computer interaction, and market research. This paper explores the efficacy of three deep learning models, YOLO5, YOLO8, and VGG16, in detecting emotions from facial features. Leveraging the capabilities of convolutional neural networks (CNNs), particularly the YOLO family known for real-time object detection, and the depth of VGG16, we aim to assess their performance in accurately localizing emotional regions and recognizing emotions. Through extensive experiments on a comprehensive dataset, comprising various emotional expressions, we evaluate the models' effectiveness and compare their performance. Additionally, we conducted a literature review to contextualize our research within the existing landscape of emotion detection from facial features. The significant gap in emotion detection from facial features lies in achieving high accuracy, particularly when dealing with obstructed images. Common issues such as poor lighting, dense shadows, and image noise pose significant barriers to clear vision, hindering the widespread deployment and effectiveness of emotion detection systems. These challenges necessitate innovative solutions to enhance the accuracy and clarity of existing emotion detection models. Our findings not only shed light on the strengths and limitations of each model but also provide insights for future advancements in emotion detection algorithms. This research contributes to the development of more accurate and robust emotion detection systems with practical applications across domains. |
|
Keywords |
Emotion detection, Facial expressions, YOLO5, YOLO8, VGG16. |
|
URL |
