Convolutional Neural Networks for Facial Expression Recognition
Keywords:
Recognition, CNN; FER2013, VGGNet, ResNet, GoogleNet, AlexNetAbstract
Human facial expressions are a kind of communication that are frequently utilised to convey emotions. People are paying more attention to facial expression recognition (FER) technology as human-computer interface technology advances. Furthermore, humans have made some progress in the field of FER. We looked at the evolution of FER in this research, including VGGNet, ResNet, GoogleNet, and AlexNet. In addition, we looked at various CNN (Convolutional Neural Network) concepts, and we chose FER2013 as the dataset to consider. FER2013 is one of the most significant databases of human faces. We also made several improvements based on the original FER methodology. The best accuracy value we got by training the FER2013 dataset in various revised techniques was 0.6424. Finally, we generated and summarised the study's progress and shortcomings. Facial Expression
Downloads
References
A. T. Lopes, E. D. Aguiar, and T. Oliveirasantos. A facialexpression recognition system using CNN. In Graphics,Patterns and Images, pages 273–280, 2015. [2] B. E. Bejnordi, J. Lin, B. Glass, M. Mullooly, G. L.Gierach, M. E. Sherman, N. Karssemeijer, J. V. D. Laak,and A. H. Beck. Deep learning-based assessment of tumorassociatedstroma for diagnosing breast cancer inhistopathology images. In IEEE International Symposiumon Biomedical Imaging, pages 929–932, 2017. [3] C. F. Bobis, R. C. Gonza´lez, J. Cancelas, I. A´ lvarez, andJ. Enguita. Face recognition using binary thresholding forfeatures extraction. In International Conference on ImageAnalysis and Processing, page 1077, 1999. [4] H. Li, H. Li, H. Li, H. Li, and H. Li. Does resnet learn goodgeneral-purpose features? In International Conference onArtificial Intelligence, Automation and ControlTechnologies, page 19, 2017. [5] I. J. Goodfellow, D. Erhan, P. L. Carrier, A. Courville, M.Mirza, B. Hamner, W. Cukierski, Y. Tang, D. Thaler, andD. H. Lee. Challenges in representation learning: A reporton three machine learning contests. Neural Netw, 64:59–63,2015. [6] M. A. Imran, M. S. U. Miah, and H. Rahman. Facerecognition using eigenfaces. ProcCvpr, 118(5):586–591,2002. [7] Shen, Dinggang, Guorong Wu, and Heung-Il Suk. “DeepLearning in Medical Image Analysis.” Annual review ofbiomedical engineering 19 (2017): 221–248. PMC. Web.25 June 2018. [8] Y. Tu, S. Li, and M. Wang. Intelligent facial expressionrecognition system r&c-fer. In Intelligent Control andAutomation, 2008. Wcica 2008. World Congress on, pages2501–2506, 2008. [9] Y. Zhang, F. Chang, L. I. Nanjun, H. Liu, and Z. Gai.Modified alexnet for dense crowd counting. (cii), 2017.
