Thesis Details

Zarovnání obrázků sportovních pozic

Bachelor's Thesis Student: Kníže Josef Academic Year: 2021/2022 Supervisor: Herout Adam, prof. Ing., Ph.D.
English title
Alignment of Sports Pose Images
Language
Czech
Abstract

The following thesis deals with image alignment based on sport pose of person in the image. The main result of this thesis is design and implementation of two systems for image alignment based on sport pose. The first system's focus is accuracy and it will be used to create dataset, which will be use to train neural network. Second system aimed at minimal hardware requirement, which was acomplished by using neural networks. Implemented neural net managed to sucessfuly align 81.98 % of images. A set of images, sorted based on the sport pose, has been created as a part of the solution.

Keywords

Image alignment, Human pose detection, Neural networks, Computer vision, Transformations regression

Department
Degree Programme
Files
Status
defended, grade C
Date
14 June 2022
Reviewer
Committee
Čadík Martin, doc. Ing., Ph.D. (DCGM FIT BUT), předseda
Bařina David, Ing., Ph.D. (DCGM FIT BUT), člen
Burget Radek, doc. Ing., Ph.D. (DIFS FIT BUT), člen
Češka Milan, doc. RNDr., Ph.D. (DITS FIT BUT), člen
Mrázek Vojtěch, Ing., Ph.D. (DCSY FIT BUT), člen
Citation
KNÍŽE, Josef. Zarovnání obrázků sportovních pozic. Brno, 2022. Bachelor's Thesis. Brno University of Technology, Faculty of Information Technology. 2022-06-14. Supervised by Herout Adam. Available from: https://www.fit.vut.cz/study/thesis/24532/
BibTeX
@bachelorsthesis{FITBT24532,
    author = "Josef Kn\'{i}\v{z}e",
    type = "Bachelor's thesis",
    title = "Zarovn\'{a}n\'{i} obr\'{a}zk\r{u} sportovn\'{i}ch pozic",
    school = "Brno University of Technology, Faculty of Information Technology",
    year = 2022,
    location = "Brno, CZ",
    language = "czech",
    url = "https://www.fit.vut.cz/study/thesis/24532/"
}
Back to top