Automatic Annotation of Tennis Action for Content-Based Retrieval by Collaborating Audio and Visual Information

  • MIYAMORI Hisashi
    Keihanna Human Info-Communication Research Center,Communications Research Laboratory

Search this article

Description

This paper proposes an automatic annotation method of tennis actions using audio and video information collaboratively. The proposed method extracts ball-hitting times clled "impact times" using audio information, and evaluates position relations between the player and the ball at the impact time to identify player's basic actions, such as forehand swing, overhead swing, etc. Simulation results show that detection rate of impact time influences recognition rate of player's basic actions. They also show that using audio information avoids some event recognition failures that cannot be averted when using only video information, demonstrating the performance and the validity of our approach.

Journal

Keywords

Details 詳細情報について

  • CRID
    1573105977154088192
  • NII Article ID
    110003223295
  • NII Book ID
    AA10826272
  • ISSN
    09168532
  • Text Lang
    en
  • Data Source
    • CiNii Articles

Report a problem

Back to top