Gradual transition detection using average frame similarity

Volkmer, T, Tahaghoghi, S and Williams, H 2004, 'Gradual transition detection using average frame similarity', in 2004 Conference on Computer Vision and Pattern Recognition Workshop. Volume 9 - Multimedia Data and Document Engineering, Washington, DC, 27 June - 2 July 2004.


Document type: Conference Paper
Collection: Conference Papers

Title Gradual transition detection using average frame similarity
Author(s) Volkmer, T
Tahaghoghi, S
Williams, H
Year 2004
Conference name International Workshop on Multimedia Data and Document Engineering
Conference location Washington, DC
Conference dates 27 June - 2 July 2004
Proceedings title 2004 Conference on Computer Vision and Pattern Recognition Workshop. Volume 9 - Multimedia Data and Document Engineering
Publisher IEEE
Place of publication Piscataway, NJ
Abstract Segmenting digital video into its constituent basic semantic entities, or shots, is an important step for effective management and retrieval of video data. Recent automated techniques for detecting transitions between shots are highly effective on abrupt transitions. However, automated detection of gradual transitions, and the precise determination of the corresponding start and end frames, remains problematic. In this paper, we present a gradual transition detection approach based on average frame similarity and adaptive thresholds. We report good detection results on the TREC video track collections - particularly for dissolves and fades - and very high accuracy in identifying transition boundaries. Our technique is a valuable new tool for transition detection.
Subjects Business Information Management (incl. Records, Knowledge and Information Management, and Intelligence)
DOI - identifier 10.1109/CVPR.2004.83
Copyright notice © 2004 IEEE
Versions
Version Filter Type
Altmetric details:
Access Statistics: 218 Abstract Views  -  Detailed Statistics
Created: Wed, 22 Jul 2009, 15:47:23 EST by Catalyst Administrator
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us