Topological mapping: a dimensionality reduction method for efficient video search

  • Authors:

  • Zaher Aghbari;Kunihiko Kaneko;Akifumi Makinouchi

  • Affiliations:

  • Kyushu University, 6-10-1 Hakozaki, Higashi-ku, Fukuoka-shi 812-8581,Japan;Kyushu University, 6-10-1 Hakozaki, Higashi-ku, Fukuoka-shi 812-8581,Japan;Kyushu University, 6-10-1 Hakozaki, Higashi-ku, Fukuoka-shi 812-8581,Japan

  • Venue:
  • Proceedings of the 2002 ACM symposium on Applied computing
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we present a novel approach for efficient search of shots by the colors of objects that exist in the wanted shots. The idea is to map the color histogram, which represents the color content, of each object in the shot from the high-dimensional feature space into a point in a low-dimensional distance space. Then, a spatial access method, such as an R-tree, is used to cluster these points based on their distances in the low-dimensional space. Our mapping method, called topological mapping, guarantees no false dismissals in the result of a query. However, the result of a query might contain some false alarms. Hence, two refinement steps are performed to remove these false alarms. Comparative experiments show the superiority in efficiency of the topological mapping method over other methods.