Link of paper: https://doi.org/10.35193/bseufbd.645138
In this study, emotion recognition process is performed by using deep learning methods for seven different facial expressions from the dataset (RidNet) which is created by using images that are publicly accessible from internet. Afterwards, transfer learning over RidNet is done with well-known convolutional neural network architectures such as AlexNet, GoogLeNet and ResNet101. Compound Facial Expressions of Emotion (CE) and Static Facial Expressions in the Wild (SFEW) datasets are determined to be used as test datasets. In the first experimental studies, convolutional neural network architecture with the best classification performance is determined. This convolutional neural network is trained using AffectNet, The Karolinska Directed Emotional Faces and RidNet. Similar classification performances are achieved when the AffectNet, KDEF, and RidNet-trained networks are tested with the dataset (CE) generated in a controlled environment. In the test dataset (SFEW) in an uncontrolled environment, RidNet-trained network gives a significant advantage over the other networks.
- MATLAB R2018b
- Deep Learning Toolbox
Please cite as:
@research article { bseufbd645138, journal = {Bilecik Şeyh Edebali Üniversitesi Fen Bilimleri Dergisi}, issn = {}, eissn = {2458-7575}, address = {Bilecik Şeyh Edebali Üniversitesi ,Fen Bilimleri Enstitüsü, Gülümbe Yerleşkesi, Bilecik}, publisher = {Bilecik Seyh Edebali University}, year = {2019}, volume = {6}, pages = {384 - 396}, doi = {10.35193/bseufbd.645138}, title = {Yeni Bir Veri Kümesi (RidNet) Kullanarak Kontrolsüz Ortamda Yüz İfadesi Tanımanın Derin Öğrenme Yöntemleri ile İyileştirilmesi}, key = {cite}, author = {Özdemi̇r, Rıdvan and Koç, Mehmet} }