Automated Content Metadata Extraction Services Based on MPEG Standards

This paper discusses the use of MPEG standards in the design and development of interoperable cloud based media architectures and services. Both offline and real-time media processing architectures are presented. Practical algorithms for metadata extraction from video content are discussed at a high level. Examples are given of derived metadata in MPEG-7 Detailed Audiovisual Profile (DAVP) for contentbased video sampling, face detection, image and face similarity, and content-based copy detection representations. A prototype system is presented in which live video encapsulated in MPEG-2 transport streams is ingested and incrementally updated MPEG-7 DAVP descriptions generated by our video analysis algorithms in real-time and delivered to clients with low delay using standard Web protocols. It is suggested that this capability can enable a range of new services such as content-based personalization of live broadcasts given that the MPEG-7 based data models fit in well with specifications for advanced television services such as TV-Anytime and ATIS IPTV Interoperability Forum.