Explanatory-description adding apparatus, computer program product, and explanatory-description adding method

摘要

When a scene group including plural scenes, for example a group of scenes bookmarked during viewing of video contents is inputted, a combination of a scene and metadata in a group of metadata that represents characteristics of the scene, corresponding to the respective scenes in the scene group, which combination has a largest distance between the metadata, is selected as explanatory descriptions that are explanations for distinguishing among the scenes, and the selected explanatory descriptions for each scene included in the scene group is added to each scene.

主权项

1. An explanatory-description adding apparatus comprising:
a collecting unit that collects the certain time associated with the scene and text information associated with the scene at the certain time; a clipping unit that clips partial text information from the text information collected by the collecting unit; an extracting unit that collectively extracts a plurality of scenes that are physically different but logically the same from among the collected scenes, as the same scenes; a weighting unit that analyzes the partial text information associated with the scenes that are collectively extracted by the extracting unit as the same scenes, and weights word information included in the partial text information for each of the scenes; an extracting unit that extracts the word information exceeding a predetermined threshold value, as the metadata of the scene associated with the word information; a storage unit that stores scene metadata as combinations of the scene at the certain time in video contents and a group of metadata representing characteristics of the scene; an input unit that inputs a scene group including a plurality of the scenes; a retrieving unit that retrieves a group of metadata corresponding to the respective scenes included in the scene group, from the scene metadata; and an adding unit that selects combinations of the metadata having a difference by deleting metadata appearing in all scenes of the scene group such that a distance among the metadata is a maximum for the scenes included in the scene group, based on a combination of the scene and the metadata retrieved by the retrieving unit, and adds the metadata to the scenes of the scene group, as explanatory descriptions, wherein, when the deleting of the metadata appearing in all scenes of the scene group deletes all of the metadata for a first scene in the scene group and a second scene in the scene group has remaining metadata, the adding unit adds metadata for the first scene as a time-series description using a temporal relationship with the second scene.