Scalable video summarization
AssigneeHewlett-Packard Development Company, L.P.
InventorsPeng Wu, Ying Li, Daniel R. Tretter
Life of this patent
Estimating the term…
Abstract
A system and method for merging scenes in a video sequence and constructing a keyframe to represent the underlying merged video content includes decomposing a video sequence into a series of component scenes, merging component scene pairs until a predetermined number of scene sets remain, extracting a keyframe from each scene set containing a single component scene, and constructing a new keyframe for each scene set containing a plurality of component scenes.
Claims24
A method for summarizing a video sequence comprising the steps of:
- decomposing a video sequence into a series of component scenes;
- merging component scene pairs until a predetermined number of scene sets remain;
- extracting a keyframe from each scene set containing a single component scene;
- and
- constructing a new keyframe for each scene set containing a plurality of component scenes.
A method to construct a new keyframe comprising the steps of:
- selecting a predetermined number of component keyframes from a video sequence;
- quantizing each of the component keyframes into one of a plurality of predefined visibility levels;
- assigning the component keyframes to grid spaces covering a predetermined new keyframe display area;
- sequentially expanding the grid spaces to a predetermined size based on the visibility level of the assigned component keyframe;
- pushing unexpanded grid spaces to avoid expanded grid spaces;
- discarding grid spaces that are pushed outside the predetermined new keyframe display area.
A method for creating a DVD title-and-chapter structure from a video sequence comprising the steps of:
- decomposing a video sequence into a series of component scenes;
- iteratively merging component scene pairs until a predetermined number of scene sets remain;
- constructing a new keyframe for each scene set having a plurality of component scenes, wherein constructing a new keyframe includes the steps of:
- selecting a plurality of component keyframes from the scene set;
- quantizing each of the plurality of component keyframes into one of a plurality of visibility levels;
- assigning the component keyframes to grid spaces covering a new keyframe display area;
- and
- sequentially expanding the grid spaces based on the visibility level of the assigned component keyframe.
A computer readable medium containing instructions for controlling a computer system to perform a method for constructing a keyframe from a scene set comprising the steps of:
- selecting a predetermined number of component keyframes from the scene set;
- quantizing each of the component keyframes into one of a plurality of predefined visibility levels;
- assigning the component keyframes to grid spaces covering a predetermined display area;
- sequentially expanding the grid spaces to a predetermined display size based on the visibility level of the assigned component keyframe;
- and
- discarding grid spaces that are expanded outside the predetermined display area.
A video summarization system comprising:
- a processor for merging component scenes of a video sequence and constructing a new keyframe for merged scenes;
- a data storage system, wherein the processor stores the video summary in the data storage system;
- and
- a display device for displaying the video summary.
Dependencies are inferred from the claim text.