The similarity of video based on the association graph construction of video objects

Abstract

We have proposed 9DST approach to represent the spatial-temporal relations between objects in a symbolic video and the similarity of videos is relevant to the users' requests. In this paper, based on the 9DST approach, we proposed the similarity retrieval algorithm. First, we construct the 9DST index structure, from the 9DST-strings, which contains the spatial-temporal relations for each pair of objects in a video database. Second, we use the similar pairs to define various types of similarity measures and construct the association graph to calculate the similarity between videos. By providing different level types of similarity between videos, our proposed similarity retrieval algorithm has discrimination power about different criteria. © 2010 Springer-Verlag Berlin Heidelberg.