Thesis Details

Rozpoznávání emocí pomocí konvolučních neuronových sítí

Bachelor's Thesis Student: Jileček Jan Academic Year: 2015/2016 Supervisor: Hradiš Michal, Ing., Ph.D.
English title
Convolutional Neural Networks for Emotion Recognition
Language
Czech
Abstract

Convolutional neural networks are used for various tasks, but foremost in machine learning, in whichthey excel. This work is going to introduce some existing frameworks, other algorithms forrecognition and then we describe the training dataset creation and the model for emotion recognitiontraining process. Mentioned model has accuracy of 60%. It is used for emotion statistics retrievalfrom movie trailers. Model for genre recognition is created from those statistics and then finally usedin our application for genre recognition of the input trailer, with best accuracy of 47%.

Keywords

Convolutional neural networks, Facial recognition, Emotion recognition, Movie genre recognition,Caffe framework, CK, AMFED, KDEF, SFEW, Brazilian FEI, K-nearest neigbours, OpenCV.

Department
Degree Programme
Information Technology
Files
Status
defended, grade C
Date
16 June 2016
Reviewer
Committee
Herout Adam, prof. Ing., Ph.D. (DCGM FIT BUT), předseda
Drahanský Martin, prof. Ing., Dipl.-Ing., Ph.D. (DITS FIT BUT), člen
Chudý Peter, doc. Ing., Ph.D. MBA (DCGM FIT BUT), člen
Očenášek Pavel, Mgr. Ing., Ph.D. (DIFS FIT BUT), člen
Strnadel Josef, Ing., Ph.D. (DCSY FIT BUT), člen
Citation
JILEČEK, Jan. Rozpoznávání emocí pomocí konvolučních neuronových sítí. Brno, 2016. Bachelor's Thesis. Brno University of Technology, Faculty of Information Technology. 2016-06-16. Supervised by Hradiš Michal. Available from: https://www.fit.vut.cz/study/thesis/18304/
BibTeX
@bachelorsthesis{FITBT18304,
    author = "Jan Jile\v{c}ek",
    type = "Bachelor's thesis",
    title = "Rozpozn\'{a}v\'{a}n\'{i} emoc\'{i} pomoc\'{i} konvolu\v{c}n\'{i}ch neuronov\'{y}ch s\'{i}t\'{i}",
    school = "Brno University of Technology, Faculty of Information Technology",
    year = 2016,
    location = "Brno, CZ",
    language = "czech",
    url = "https://www.fit.vut.cz/study/thesis/18304/"
}
Back to top