Method of Extracting Shape Variation Descriptor for Retrieving Image Sequence

摘要

PURPOSE: A method for extracting a shape variation descriptor for searching image sequence is provided to accurately search image sequence requested by a user, by identifying image sequence data in which only a part of an object moves or the object has partial shape variation. CONSTITUTION: A method for extracting a shape variation descriptor for searching image sequence includes the steps of selecting a predetermined number of frames from image sequence data(S101); transforming the frames into frames including only object information divided from a background(S103); arranging an object at the predetermined positions of the frames(S105); generating one frame including motion information and shape information of the object by overlapping the arranged frames(S107); and extracting a shape descriptor to the generated frame(S109).