Please use this identifier to cite or link to this item:
Title: Semantic soccer video analysis
Keywords: mid-level representation, semantic descriptor
Issue Date: 27-Aug-2004
Citation: SUN HAIPING (2004-08-27). Semantic soccer video analysis. ScholarBank@NUS Repository.
Abstract: This thesis describes a study on a novel intermediate representation to bridge the gap between low-level features and semantic meanings in soccer video analysis. Based on our study, we conclude that shot is not suitable as a mid-level representation for soccer video parsing. We classify a soccer video into seven categories, and assign each category a word, called semantic descriptor, to indicate its semantic meaning. Therefore, a soccer video can be represented by a semantic descriptor sequence. Two different approaches, segment-based and frame-based approaches, were designed to realize this representation A total of 450 minutes of soccer video from FIFA2002 were used to test the two approaches. The average accuracy is 81.0% for the frame-based method and 76.1% from the segment-based.
Appears in Collections:Master's Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Thesis_Semantic Soccer Video Analysis.pdf1.12 MBAdobe PDF



Page view(s)

checked on Oct 7, 2018


checked on Oct 7, 2018

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.