Affective computing for visual emotion recognition using convolutional neural networks

• Main Authors: Ashraf, Arselan, Gunawan, Teddy Surya, Sophian, Ali, Ambikairajah, Eliathamby, Ihsanto, Eko, Kartiwi, Mira
• Format: Book Chapter
• Language: English; English; English
• Published: Springer 2021
• Subjects: TK7885 Computer engineering
• Online Access: http://irep.iium.edu.my/86114/1/86114_Acceptance%20letter.pdf; http://irep.iium.edu.my/86114/8/86114_Affective%20computing%20for%20visual%20emotion%20recognition.pdf; http://irep.iium.edu.my/86114/14/86114_Affective%20computing%20for%20visual%20emotion%20recognition-scopus.pdf; http://irep.iium.edu.my/86114/; https://icites2020.ump.edu.my/index.php/en/
• Institution: Universiti Islam Antarabangsa Malaysia

Affective computing is a developing interdisciplinary examination field uniting specialists and experts from different fields, going from artificial intelligence, nat-ural language processing, to intellectual and sociologies. The thought behind Af-fective Computing is to give PCs the aptitude of insight that will, in general, comprehend human feelings. Notwithstanding, these victories, the field needs hypothetical firm establishments and efficient rules in numerous regions, espe-cially so in feeling demonstrating and the development of computational models of feeling. This exploration manages Affective Computing to improve the exhibi-tion of Human-Machine Interaction. The focal point of this work is to distinguish the emotional state of a human utilizing deep learning procedure, i.e. Convolu-tional Neural Networks (CNN). The Warsaw Set of Emotional Facial Expression Pictures dataset has been utilized to build up a feeling acknowledgement model which will have the option to perceive five facial feelings, including happy, sad, anger, surprise and neutral. The proposed framework design and the strategy has been discussed in this paper alongside the experimental findings.