Robust Visual Content Representation Using Compression Modes Driven Low-level Visual Descriptors
In conventional visual content representation, low-level visual features are usually extracted from the highest quality and resolutions of visual contents. When visual content is scalable coded and utilised, their bit streams can be adapted at various nodes in multimedia usage chains to cater the variations in network bandwidths, display device resolutions and resources and usage preferences by just discarding insignificant resolution-quality layers. This can result in the existence of different version of the same content with dissimilar low-level visual features. Therefore, mapping of low level visual descriptors into content resolution-quality spaces is important in order to obtain low-level visual features that are robust to such content adaptations. A new scalable domain feature extraction using the compression modes and decisions is presented and its content based image retrieval performance is evaluated. The proposed scheme outperforms MPEG-7 visual descriptors in both the original image and scaled resolution-quality space domains.
Keywordscontent representation low-level descriptors scalable coding content adaptation MPEG-7 CBIR wavelets EZW
Unable to display preview. Download preview PDF.
- 2.ITU-T, JTC1, I.: Advanced video coding for generic audiovisual services. ITU-T Rec. ISO/IEC 14496-10 (2003)Google Scholar
- 4.Manjunath, B.S., Salembier, P.P., Sikora, T.: Introduction to MPEG-7: Multimedia Content Description Interface. John Wiley and Sons, Chichester (2002)Google Scholar
- 5.Lu, Z.-M., Li, S.-Z., Burkhardt, H.: A content-based image retrieval scheme in jpeg compressed domain. Int’l. Jnl. of Innovative Computing, Information and Contorl, 831–839 (2005)Google Scholar
- 9.Messing, D.S., Van Beek, P., Errico, J.H.: The mpeg-7 colour structure descriptor: Image description using colour and local spatial information. In: Proc. IEEE ICIP, pp. 670–673 (2001)Google Scholar
- 10.Jasutani, E., Yamada, A.: The mpeg-7 colour layout descriptor: A compact feature description for high speed image/video segment retrieval. In: Proc. IEEE ICIP, pp. 674–677 (2001)Google Scholar
- 11.Li, Y., Shapiro, L.: Object and concept recognition for cbir image and ground truth database, http://www.cs.washington.edu/research/imagedatabase/groundtruth/