Jeong Mi Young, Lee Won Suk

The Transactions of the Korea Information Processing Society (1994 ~ 2000), Vol. 4, No. 4, pp. 908-922, Apr. 1997
10.3745/KIPSTE.1997.4.4.908,   PDF Download:

Abstract

In this paper, we present an efficient way to describe the complex meaning embedded in video data for the content-based retrieval in a video database. Instead of viewing the data stored in a video file as a continuous bit stream, we associate the desired interval in the stream with a set of attribute and value pairs which describe its meaning. Since the meaning of video data can be complex and can change dynamically, it is necessary to combine the fixed schema of attributes used in the traditional database system with a description method through a set of attribute and value pairs that can be defined dynamically. As the content of video data can be expressed differently according to the view point of a user, it is important to maintain the meaning of the set of attribute and value pairs. In addition, it also describes a way to define a new video presentation by separating a video stream physically or by sharing the portion of a bit stream, and the new method can minimize the required storage space.

Statistics
Cite this article