The Community for Technology Leaders
Green Image
We introduce a new data model called algebraic video that provides operations for the composition, search, navigation and playback of digital video presentations. Video presentations are composed using a video algebra that consists of a set of basic operations on video segments to produce a desired video stream. The video algebra contains operations for temporally and spatially combining video segments as well as for attaching attributes to these segments. Algebraic video access methods also query and navigation operations. Query and navigation allow users to discover video presentations of interest by describing desired attributes and exploring a presentation's context. Unlike previous approaches, algebraic video permits video expressions to be nested in arbitrarily deep hierarchies. It also permits video segments to inherit attributes by context. Experience with a prototype algebraic video system suggests that algebraic video is an effective way to access and manage video. The prototype system is used to discover video segments of interest from existing collections and create new video presentations with algebraic combinations of these segments.
algebraic video, video algebra, content-based retrieval, video indexing and searching, video databases

R. Weiss, A. Duda and D. K. Gifford, "Composition and Search with a Video Algebra," in IEEE MultiMedia, vol. 2, no. , pp. 12-25, 1995.
103 ms
(Ver 3.3 (11022016))