Building a Large Annotation Ontology for Movie Video Retrieval

Multimedia content continues to grow rapidly. To ensure access to growing video collections, semantic indexing of images and videos has become a very important issue for data access, retrieval and actual application. For developing and evaluating semantic concepts searching annotation techniques, it is necessary to predefine a large lexicon, construction of a large benchmark data set, and annotation of videos in a rigorous fashion. In this paper we developed a method of movie's video benchmark data set, which includes (1) design of structure of movie semantic annotation ontology, (2) definition of lexicons and concepts that accommodate consumers' needs, and (3) building semantic relations among lexicons and concepts. To our knowledge, this is the first systematic work in the movie domain aimed at the definition of a large lexicon, construction of a large benchmark data set, and annotation of videos in a rigorous fashion. We test performance of the ontology on movie's video, and experiments demonstrate that our method is capable of annotation for movie's video.