Unsupervised summarization of rushes videos

Yang Liu, Feng Zhou, Wei Liu, Fernando De La Torre, Yan Liu

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

13 Citations (Scopus)

Abstract

This paper proposes a new framework to formulate summarization of rushes video as an unsupervised learning problem. We pose the problem of video summarization as one of time-series clustering, and proposed Constrained Aligned Cluster Analysis (CACA). CACA combines kernel k-means, Dynamic Time Alignment Kernel (DTAK), and unlike previous work, CACA jointly optimizes video segmentation and shot clustering. CACA is effciently solved via dynamic programming. Experimental results on the TRECVID 2007 and 2008 BBC rushes video summarization databases validate the accuracy and effectiveness of CACA.

Original languageEnglish
Title of host publicationMM'10 - Proceedings of the ACM Multimedia 2010 International Conference
Pages751-754
Number of pages4
DOIs
Publication statusPublished - 1 Dec 2010
Event18th ACM International Conference on Multimedia ACM Multimedia 2010, MM'10 - Firenze, Italy
Duration: 25 Oct 201029 Oct 2010

Publication series

NameMM'10 - Proceedings of the ACM Multimedia 2010 International Conference

Conference

Conference18th ACM International Conference on Multimedia ACM Multimedia 2010, MM'10
Country/TerritoryItaly
CityFirenze
Period25/10/1029/10/10

Keywords

  • constrained aligned cluster analysis
  • dynamic programming
  • dynamic time alignment kernel
  • rushes video summarization
  • trecvid
  • unsupervised learning

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Human-Computer Interaction
  • Software

Cite this