Abstract
A content-based video scene retrieval method using a multimodal index is proposed. Representative images of video scenes and corresponding visualized sound patterns are used as the multimodal index. Color-coded patterns of the sound spectrogram are adopted as the sound index. An image search engine is used not only for image retrieval but also for sound pattern retrieval. The results of parallel query experiments using the multimodal index suggest that it is more effective in improving query accuracy than single query methods. The results of video index browsing experiments indicate the effectiveness of the image-sound combined index method for efficient video queries and understanding the content.
Chapter PDF
Similar content being viewed by others
References
Aigrain, P., Zhang, H. and Petkovic, D. (1996) Content-based representation and retrieval of visual media: A state-of-the-art review, Multimedia Tools and Applications 3, 179–202.
Brown, M. G., Foote, J. T., Jones, G. J. F. Jones, K. S., and Young, S. J. (1996) Open-vocabulary speech Indexing for voice and video mail retrieval, Proc. ACM Multimedia 96, 307–316, Boston, ACM.
Fushikida, K., Hiwatari, Y. and Waki, H. (1998) Content-based image query method using parallel retrieval scheme, ICCIMA 98, 830–835.
Hauptmann, A.G. and Smith, M. (1995) Text Speech, and Vision for Video Segmentation: The Informedia Project, AAA’ Fall 1995 Symposium on Computational Models for Integrating Language and Vision.
Hirata, Y., Hara, K., Shibata, N. and Hirabayashi, F. (1993) Media-based Navigation for Hypermedia system, ACM Hypertext, 157–173.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1998 Springer Science+Business Media New York
About this chapter
Cite this chapter
Fushikida, K., Hiwatari, Y., Waki, H. (1998). A Content-Based Video Retrieval Method Using a Visualized Sound Pattern. In: Ioannidis, Y., Klas, W. (eds) Visual Database Systems 4 (VDB4). VDB 1998. IFIP — The International Federation for Information Processing. Springer, Boston, MA. https://doi.org/10.1007/978-0-387-35372-2_18
Download citation
DOI: https://doi.org/10.1007/978-0-387-35372-2_18
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4757-6939-5
Online ISBN: 978-0-387-35372-2
eBook Packages: Springer Book Archive