In this paper we have proposed a scheme for parsing News video sequences into their semantic components using integrated aural and visual features. We have explored use of the Token Passing Algorithm with HMM for simultaneous segmentation and characterization of the components. Experimentation with about 100 sequences have shown impressive results. © Springer-Verlag Berlin Heidelberg 2005.
CITATION STYLE
Krishna, K. K., Subbarao, R., Chaudhury, S., & Kumar, A. (2005). Parsing news video using integrated audio-video features. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 3776 LNCS, pp. 538–543). https://doi.org/10.1007/11590316_85
Mendeley helps you to discover research relevant for your work.