Thesis Details

Optical Flow Methods for Video and Image Segmentation

Bachelor's Thesis Student: Tabernero Diego Academic Year: 2018/2019 Supervisor: Beran Vítězslav, doc. Ing., Ph.D.
Czech title
Optical Flow Methods for Video and Image Segmentation
Language
English
Abstract

This work presents a method to detect stable, or almost stable video-parts from the original video footage with large amount of fast large transformations (zooming, translation, rotation). It deals with approaches based on optical flow for getting information about the feature vectors of the video in each transition between frames. The classification methods used to determine if the consecutive frames are stable or not are machine learning methods like support vector machine (SVM) and K-means.

Keywords

Optical flow, machine learning, Lucas Kanade, Gunnar Fanerbäck, SVM, K-means

Department
Department of Computer Graphics and Multimedia FIT BUT
Degree Programme
Shortterm study BSc., Field of Study Information Technology
Files
Status
defended, grade C
Date
12 June 2019
Reviewer
Španěl Michal, Ing., Ph.D.
Citation
TABERNERO, Diego. Optical Flow Methods for Video and Image Segmentation. Brno, 2019. Bachelor's Thesis. Brno University of Technology, Faculty of Information Technology. 2019-06-12. Supervised by Beran Vítězslav. Available from: https://www.fit.vut.cz/study/thesis/22103/
BibTeX 
@bachelorsthesis{FITBT22103,
    author = "Diego Tabernero",
    type = "Bachelor's thesis",
    title = "Optical Flow Methods for Video and Image Segmentation",
    school = "Brno University of Technology, Faculty of Information Technology",
    year = 2019,
    location = "Brno, CZ",
    language = "english",
    url = "https://www.fit.vut.cz/study/thesis/22103/"
}

Theses

Back to top