Show simple item record

dc.rights.licenseCC-BY-NC-ND
dc.contributor.advisorVeltkamp, R.C.
dc.contributor.authorBroeder, S.A.A. den
dc.date.accessioned2019-03-22T18:00:08Z
dc.date.available2019-03-22T18:00:08Z
dc.date.issued2019
dc.identifier.urihttps://studenttheses.uu.nl/handle/20.500.12932/31995
dc.description.abstractIn recent years there has been an increase in published research on video games. However, few articles discuss semantic analysis of gameplay videos, which are available online on platforms such as YouTube or Twitch. A case-study was performed where a three-step approach is proposed to analyze a specific scene in videos of the game ‘This War of Mine’. The first step is to detect the location at which the scene of interest takes place using a SVM with bag-of-visual-words histograms of SIFT features. Then convolutional neural networks are used to detect which scene takes place at that location and what choice the player makes during that scene.
dc.description.sponsorshipUtrecht University
dc.format.extent491124
dc.format.mimetypeapplication/pdf
dc.language.isoen_US
dc.titleAutomatic semantic analysis of gameplay videos of ‘This War of Mine’
dc.type.contentMaster Thesis
dc.rights.accessrightsOpen Access
dc.subject.keywordsComputer vision, video analysis, game analysis, machine learning
dc.subject.courseuuGame and Media Technology


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record