QUALITATIVE CHARACTERIZATION OF DYNAMIC TEXTURES FOR VIDEO RETRIEVAL

Abstract

A new issue in texture analysis is its extension to temporal domain, known as dynamic texture. Many real-world textures are dynamic textures whose retrieval from a video database should be based on both dynamic and static features. In this article, a method for extracting features revealing fundamental properties of dynamic textures is presented. Their interpretation enables qualitative requests when browsing videos. Future work is finally exposed.