Behavior Classification Using Speed Invariant Feature

DOI DOI Web Site
  • KUNO Tomonori
    Faculty of Informatics and Engineering, The University of Electro-Communications
  • TAKAHASHI Hiroki
    Graduate School of Informatics and Engineering, The University of Electro-Communications

Bibliographic Information

Other Title
  • 速度不変特徴量を用いた動作識別(画像処理・感性,映像表現・芸術科学フォーラム2016)
  • 速度不変特徴量を用いた動作識別
  • ソクド フヘン トクチョウリョウ オ モチイタ ドウサ シキベツ

Search this article

Abstract

Recently, security camera has been widely used in security system. Some method for behavior classification are also interesting topic for many researchers, CHLAC and subspace are method used in behavior classification. However, this method are not able to classify the behavior of the same type of motion occurred in different speeds. Therefore, this paper proposes a different approach that considers the speed invariant and perform behavior classification without depending on the motion speed. The experiment is carried out using seven types of motion behavior in three different speeds. Finally, evaluation is conducted for the effectiveness.

Journal

  • ITE Technical Report

    ITE Technical Report 40.11 (0), 25-28, 2016

    The Institute of Image Information and Television Engineers

Keywords

Details 詳細情報について

Report a problem

Back to top