A web-enabled video indexing system

  • Authors:

  • Jian Zhou;Xiao-Ping Zhang

  • Affiliations:

  • Ryerson University, Toronto, Ontario, Canada;Ryerson University, Toronto, Ontario, Canada

  • Venue:
  • Proceedings of the 6th ACM SIGMM international workshop on Multimedia information retrieval
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Video parsing and indexing is an important early stage of content-based video analysis. In this paper, we present a new web-enabled video indexing system that integrates Synchronized Multimedia Integration Language (SMIL) standard. New algorithms are proposed for video temporal segmentation. Sharp transition detection is achieved by an enhanced histogram-based method that is robust to illumination changes. For gradual transition detection, new features are introduced for dissolve detection. The proposed dissolve detector is based on a combined analysis of mean-variance-skewness of intensity. Compared with existing variance-based approaches, the introduced new features improve the discrimination ability on shot boundaries. We also describe methods for eliminating false positives. Experimental results show that the proposed algorithms can effectively detect shot boundaries. Detected scenes and other cinematic attributes are structured and organized by integrating HTML and SMIL. For each video file, the system generates a table-of-contents indexing file. The user-friendly interface provides web-based interaction, browsing and previewing of video content