TY - CHAP T1 - Toward motion picture grammars T2 - Computer Vision — ACCV'98Computer Vision — ACCV'98 Y1 - 1997 A1 - Bolle,Ruud A1 - Aloimonos, J. A1 - Fermüller, Cornelia ED - Chin,Roland ED - Pong,Ting-Chuen AB - We are interested in processing video data for the purpose of solving a variety of problems in video search, analysis, indexing, browsing and compression. Instead of concentrating on a particular problem, in this paper we present a framework for developing video applications. Our basic thesis is that video data can be represented at a higher level of abstraction as a string generated by a grammar, termed motion picture grammar. The rules of that grammar relate different spatiotemporal representations of the video content and, in particular, representations of action. JA - Computer Vision — ACCV'98Computer Vision — ACCV'98 T3 - Lecture Notes in Computer Science PB - Springer Berlin / Heidelberg VL - 1352 SN - 978-3-540-63931-2 UR - http://dx.doi.org/10.1007/3-540-63931-4_228 ER -