Content Based Multi-Media Analytics (CBMA)

This report provides a summary of the research developments of IST-144-RTG on Content BasedMulti-Media Analytics (CBA) carried out by a team from NLD, NOR, USA, and UK. Thesescientists have brought together information retrieval strategies from heterogeneous media sources(text, video and images) and human assessment. As a result, multiple heterogeneous data sourcescan be exploited by content based information retrieval and multimedia analytics to deliver timelyand accurate synopses of data with information that can be combined with human intuition andunderstanding to develop a comprehensive view of the problem/solution space. Such interoperabletools that cross-cue knowledge obtained from one method to generate taskings in another are neededby NATO Coalition military leaders, commanders, and intelligence analysts to accelerate situationalawareness and decision making and deal with the complexity of the defence information space.A description of the core technical components is provided along with the description of theircombined application in a concept demonstrator addressing a fictitious, but nevertheless realistic,scenario representative of defence challenges facing the alliance. The report is concluded with thefindings and recommendations for further work that is needed to address limitations, includingtechnical and system level gaps, such as an open distributed architecture for integration of analyticsservices, now and expected in the future.