Digital video managing and searching system

摘要

A system for managing and searching for a digital video includes: a video feature point extraction unit decoding an input video and extracting a feature point; a video feature point database (DB) storing and managing a feature point of a video to be compared (i.e., a comparison target video), and a video feature point comparison unit coarsely comparing the feature point of the input video and that of the comparison target video to acquire a candidate group, and minutely comparing the candidate group to detect a content repeated section, and informing a user of the content repeated section.

主权项

1. A system for managing and searching for a digital video, the system comprising:
a video feature point extraction unit decoding an input video and extracting a feature point; a video feature point database (DB) storing and managing a feature point of a comparison target video; and a video feature point comparison unit coarsely comparing the feature point of the input video and that of the comparison target video to acquire a candidate group, and minutely comparing the candidate group to detect a content repeated section, and informing a user of the content repeated section, wherein the video feature point extraction unit comprises: a coarse feature extractor receiving input video and outputting coarse features; a fine feature extractor receiving the input video and outputting fine features; and a feature point structure generator generating feature points from the coarse features and the fine features, wherein the feature point structure generator comprises: a feature merger merging frames based on the coarse features; an index generation arranger extracting an index of the frames which have been merged by the feature merger; and a fine feature organizer organizing the number of fine features to be included in the feature points according to the number of the frames which have been merged by the feature merger, wherein the index generation arranger comprises: an index extractor selecting bits or values resistant to various video deformation from the video feature points extracted by the video feature extractor and combining them to configure information regarding the index; an entire index organizer extracting information regarding every frame from the video feature points extracted by the video feature extractor to generate an entire frame index table; and a key frame index organizer extracting position information of a key frame with respect to every frame from the video feature points extracted by the video feature extractor to generate a key frame index table, wherein the video feature point comparator comprises: an index comparator comparing the feature points of the input video an those of the comparison target video to recognize feature points having the same index; a candidate detector detecting a candidate group by connecting the feature points; and a repeated section detector analyzing fine features of the candidate group to detect a section of the input video having the same content as that of the comparison target video.