Scene change detection in compressed video data

Date

1995-08

Journal Title

Journal ISSN

Volume Title

Publisher

Texas Tech University

Abstract

Use of digital video data is on the rise in applications like multimedia educational and training tools, multimedia mail, networked video conferencing systems and desktop entertainment systems.

Due to the large amount of data associated with video applications, video data is compressed before storage or transmission. Therefore the problems of content based indexing and retrieval of compressed video data assume primary importance. Solutions to these problems require computationally efficient detection of scene changes. Previous research has shown that scene change techniques operating on compressed data are much faster than techniques requiring decompression of data before scene change detection.

We have developed new scene change detection algorithms, based on block comparisons between adjacent frames of digital video data compressed in the spatial frequency domain.

Description

Citation