News story segmentation in multiple modalities

In this paper, we describe an approach to segmenting news video based on the perceived shift in content using features spanning multiple modalities. We investigate a number of multimedia features, which serve as potential indicators of a change in story in order to determine which are the most effective. The efficacy of our approach is demonstrated by the performance of our prototype, where a number of feature combinations demonstrate an up to 18% improvement in WindowDiff score above that of other state of the art story segmenters. In our investigation, there was no, one, clearly superior feature, rather the best segmentation results occurred when there was synergy between multiple features.

• Type: Journal
• Year: 2010
• Where: MTA
• Authors: Gert-Jan Poulisse, Marie-Francine Moens, Tomas Dekens, Koen Deschacht
• Keywords: Art Story Segmenters, MTA 2010, Multiple Modalities, Potential Indicators