Identifying Spatial and Temporal Media Fragments on the Web

摘要: Semantic descriptions of non-textual media available on the Web can facilitate retrieval and presentation assets documents that contain them. languages represent controlled vocabularies shared annotations content Web. By identifying concepts to consider, Uniform Resource Identifiers (URIs) are building blocks RDF subject-predicate-object triples provide mortar by specifying relations between Often, particular regions an image or sequences a video need be localized (anchor value in [1]) uniquely identified order used as subject object resource annotation. However, current architecture does not means for sub-parts assets, same way fragment identifier URI refer part HTML XML document. Actually, almost all other types, semantics has been defined is commonly accepted. The specification defines general meaning scheme#fragment, and, example, when scheme http, RFC2616 specifies HTTP GET performed find out what is, yielding certain Content-Type response (i.e. mime-type fragment). mimetype registry within document depending its type. For ’text/html’ RFC2854 actually anchor. Providing agreed upon localize multimedia objects (e.g. sub-regions images, temporal videos tracking moving space time) fundamental [2, 3, 4, 5, 6]. requirements expressing processing these fragments have studied [7]. This position paper describes several ways using W3C recommendations, ISO standards RFC.