I.R.I.S. Working Group

Interoperable Multimedia Retrieval in Distributed Systems

Media Annotations Working Group (MAWG)

(The following text has been taken from the group charter)


This Group is chartered to provide a simple ontology to support cross-community data integration of information related to media objects on the Web, as well as an API to access the information. This ontology would help circumventing the current profileration of video metadata formats by providing full or partial translation and mapping between the existing formats.

Anticipating the increase in online video and audio in the upcoming years, it will become progressively more difficult for viewers to find the content using current search tools. Unlike hypertext documents, it is more complex and sometimes impossible to deduce meta information about a medium, such as its title, author, or creation date. An ontology would provide a common set of terms to define the basic metadata needed for media objects. There has been a proliferation of media metadata formats. For example, an image could potentially contain EXIF, IPTC and XMP information. There are several metadata solutions for media related content, including MPEG-7, IPTC, iTunes XML, Yahoo! MediaRSS, Video sitemaps, CableLabs VOD Metadata Content, TV-ANytime (ETSI TS 102 822 series), EBU Core Metadata Set, and XMP (see also Multimedia Vocabularies on the Semantic Web and the report from the Video on the Web workshop). The goal of this Group is to develop a simple lingua franca common ontology between the existing standards, and define a roadmap for future extension of the ontology. It is not expected that the Group will focus on 'low level' metadata in its first phase. The intent is to help publishers to harmonize metadata information encoded in different languages but also to help developers with the lack of syntactic and semantic interoperability. Ultimately, users of the ontology should also be able to take advantage of Semantic Web technologies, such as the SPARQL Query Language for RDF. The Group should look at the documents produced by the W3C Multimedia Semantics Incubator Group as a starting point and survey the existing metadata formats to focus on a specific set. The Group will start by focusing on discrete/finite simple video objects and will then expand at a later stage into streamed video, image, and audio.

The Group should explore a minimal approach for specifying proper copyright licensing, which may require separate terms for specifying the holder of the copyright and the licensing terms.

The Group will also develop a specification for an application programming interface (API) to access metadata information. At the minimum, the Group will provide a simple client-side readonly ECMAScript API. The Group should also consider making the underlying media metadata available to a browser ECMAScript API in some simple RDF form, especially as a number of the underlying data formats permit full RDF. Such access should be able to operate on metadata elements defined in future versions of the simple common ontology; that is, it should not be designed such that extensions are required when the common ontology is extended.

The Group may hold Workshops, Interoperability Meetings, and other events as required to fulfill its mission.

  • Abilitity to convert core metadata information from one metadata standard to an other
  • Development of a set of requirements and competency questions to determine the scope of the ontology and to test the fulfillment of the requirements and coverage of the ontology.
  • Production of stable documents addressing the work items listed in the Deliverables section.
  • Availability of at least two implementations mapping the ontology developed by the Group with two other existing metadata standards, as demonstrated by an implementation report (summarizing implementation status against the relevant test suite) for each testable class of product, including user agents
  • Availability of at least two implementations for the API specification(s).
  • Demonstrate that the 'most commonly used' metadata items in use by the industry are mapped by the ontology.

Full coverage of all metadata elements in EXIF, IPTC, XMP, MPEG-7, and similar broad vocabularies, is out of scope, as well as signal related descriptors.