Thesis Details

Rekonstrukce nekvalitních snímků obličejů

Master's Thesis Student: Bako Matúš Academic Year: 2019/2020 Supervisor: Hradiš Michal, Ing., Ph.D.
Language
Slovak
Abstract

 In this thesis, I tackle the problem of facial image super-resolution using convolutional neural networks with focus on preserving identity. I propose a method consisting of DPNet architecture and training algorithm based on state-of-the-art super-resolution solutions. The model of DPNet architecture is trained on Flickr-Faces-HQ dataset, where I achieve SSIM value 0.856 while expanding the image to four times the size. Residual channel attention network, which is one of the best and latest architectures, achieves SSIM value 0.858. While training models using adversarial loss, I encountered problems with artifacts. I experiment with various methods trying to remove appearing artefacts, which weren't successful so far. To compare quality assessment with human perception, I acquired image sequences sorted by percieved quality. Results show, that quality of proposed neural network trained using absolute loss approaches state-of-the-art methods.

Keywords

super-resolution, computer vision, neural networks, image reconstruction

Department
Degree Programme
Information Technology, Field of Study Intelligent Systems
Files
Status
defended, grade B
Date
14 July 2020
Reviewer
Committee
Rogalewicz Adam, doc. Mgr., Ph.D. (DITS FIT BUT), předseda
Bidlo Michal, doc. Ing., Ph.D. (DCSY FIT BUT), člen
Hradiš Michal, Ing., Ph.D. (DCGM FIT BUT), člen
Hrubý Martin, Ing., Ph.D. (DITS FIT BUT), člen
Rozman Jaroslav, Ing., Ph.D. (DITS FIT BUT), člen
Citation
BAKO, Matúš. Rekonstrukce nekvalitních snímků obličejů. Brno, 2020. Master's Thesis. Brno University of Technology, Faculty of Information Technology. 2020-07-14. Supervised by Hradiš Michal. Available from: https://www.fit.vut.cz/study/thesis/23003/
BibTeX
@mastersthesis{FITMT23003,
    author = "Mat\'{u}\v{s} Bako",
    type = "Master's thesis",
    title = "Rekonstrukce nekvalitn\'{i}ch sn\'{i}mk\r{u} obli\v{c}ej\r{u}",
    school = "Brno University of Technology, Faculty of Information Technology",
    year = 2020,
    location = "Brno, CZ",
    language = "slovak",
    url = "https://www.fit.vut.cz/study/thesis/23003/"
}
Back to top