Shot Change Detection Based on the Reynolds Transport Theorem

  • Authors:
  • C. C. Shih;Hsiao-Rong Tyan;Hong-Yuan Mark Liao

  • Affiliations:
  • -;-;-

  • Venue:
  • PCM '01 Proceedings of the Second IEEE Pacific Rim Conference on Multimedia: Advances in Multimedia Information Processing
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

Most of the conventional gradual shot change detection algorithms are heuristics-based. Usually, a general shot transition detection algorithm which can deal with different kinds of situations may be very tedious and complicated if it is heuristics-based. Under the circumstances, a general detection algorithm which has a solid theoretic ground is always preferrable. In this paper, we propose a general shot change detection algorithm which is able to achieve the above mentioned goal. The proposed algorithm consists of two major stages: the modeling stage and the detection stage. In the modeling stage, we model a shot transition by calculating the change of color/intensity distribution corresponding to the shots before and after a transition. In the detection stage, we consider a video clip as a continuous "frame flow" and then apply the Reynolds Transport Theorem to analyze the flow change within a pre-determined control volume. Using the above mentioned methodology, the shot change detection problem becomes theoretically analytic. Experimental results have proven the superiority of the proposed method.