Digital video scenes identification using audiovisual features

  • Authors:
  • Danilo Barbosa Coimbra;Rudinei Goularte

  • Affiliations:
  • Universidade de São Paulo, São Carlos, SP -- Brasil;Universidade de São Paulo, São Carlos, SP -- Brasil

  • Venue:
  • WebMedia '09 Proceedings of the XV Brazilian Symposium on Multimedia and the Web
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper proposes a new technique to segment digital video into semantic units called scenes. This technique is derived from the combined use of color histograms and silence detection, visual and aural features, respectively. The category of digital video used in this work is television news. The results show that the technique can identify more than 80% of the video scenes.