This paper describes AgileRAI, a framework for searching, organising and accessing multimedia data in a fast and semantic-driven way.
AgileRAI supports real-time ingestion of video streams on which different machine learning techniques (such as global and local visual features extraction and matching) are applied in a parallel and scalable way. Extracted features are matched to a reference database of visual patterns (e.g. faces, logos, monuments) in order to produce a set of meta-tags describing the ingested contents.
Furthermore, these tags are semantically enriched using open semantic data repositories. The system is designed with a scale-out pattern architecture based on Apache Spark, ensuring high-performance in Big Data management environments.
In today’s digital age, television content life cycle has a very long span: after being produced and broadcast, a copy of the content, possibly enriched with metadata, is stored in the archive to be reused when needed or to be published online.
In order to maximise the reuse of those assets, the ability to efficiently search, organise and access content in a fast and semantic-driven way is an asset of fundamental importance for the broadcast and media industry. However, efficient access to large-scale video assets requires content to be conveniently annotated through an intellectually expensive and time-consuming process.
Semantic Web technologies can provide solutions for enriching media content and improving search as they are designed to facilitate data integration and linking over a large number of data sources. Some broadcasters already started manually tagging contents, using DBpedia as a source of tag identifiers.
However, given the evergrowing amount of content produced every day and available from the archive, in order to make the process scalable, a way to automatically assign tags is still needed.