conference paper

Conference/ProceedingsProceedings of the 2nd International Conference on Semantics And digital Media Technologies (SAMT'07)
Start date05.12.2007
End date07.12.2007
Author(s)J. Nemrava, P. Buitelaar, T. Declerck, V. Svátek, J. Petrák, A. Cobet, H. Zeiner, D. Sadlier, N. O’Connor
TitleEnhancing Video Analysis Results using Complementary Textual Resources
AbstractIn this document we deal with different sources of information complementary to audio-visual streams and propose their usage for enriching this data with semantic concepts in order to bridge the gap between low-level video analysis and high-level analysis. Our aim is to extract cross-media feature descriptors from semantically enriched and aligned resources so as to detect finer-grained events in video. We introduce an
architecture for complementary resources analysis and discuss domain dependency aspects of this approach connected to our initial domain of football broadcasts.
Key wordsow-level, high-level, cross-media feature descriptors, text detection, video OCR