Significant Scene Extraction Method Using Situation Importance

  • Authors:
  • Takako Hashimoto;Takashi Katooka;Atsushi Iizawa;Hiroyuki Kitagawa

  • Affiliations:
  • -;-;-;-

  • Venue:
  • ICDCSW '04 Proceedings of the 24th International Conference on Distributed Computing Systems Workshops - W7: EC (ICDCSW'04) - Volume 7
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Recently, Sports videos for mobile terminals and PCsand other multimedia information services such as sportsvideos for mobile terminals and PCs have recentlyspread. In these services, The ability to efficientlytextraction of significant scenes becomes importantisimportant with these services. We proposed the PersonalDigest Making Scheme (PDMS), which calculates thesignificance of events in the a sports game based on videometa data and extracts significant scenes to compose adigest. PDMS calculations focus on additional-pointevents' significance, so the system could cannot analyzelow-score games correctly. To address the problem, thispaper explains the a method of calculating situationimportance before events happen, to take into accountsignificant scenes extractiowhich will be taken intoaccount when extracting significant scenesn. Situationimportance presents affords a degree of a chance. Usingsituation importance makes it possible to extractsignificant scenes more correctly.