Method for representing and comparing multimedia content according to rank
View Patent ↗A method for generating a representation of multimedia content by first segmenting the multimedia content spatially and temporally to extract objects. Feature extraction is applied to the objects to produce semantic and syntactic attributes, relations, and a containment set of content entities. The content entities are coded to produce directed acyclic graphs of the content entities, where each directed acyclic graph represents a particular interpretation of the multimedia content. Attributes of each content entity are measured and the measured attributes are assigned to each corresponding content entity in the directed acyclic graphs to rank order the multimedia content.
1. A computer implemented method for ordering multimedia content, comprising the steps of:
segmenting the multimedia content to extract video objects, in which the objects are video object planes;
extracting and associating features of the video object to produce content entities, wherein the content entities are recursive data structures comprising features, relations, directed acyclic graphs and containment sets;
coding the content entities to produce directed acyclic graphs of the content entities, each directed acyclic graph representing a particular interpretation of the multimedia content;
measuring high- level temporal attributes of each content entity;
assigning the measured high- level temporal attributes to each corresponding content entity in the directed acyclic graphs to order the content entities of the multimedia content;
comparing the ordered content entities in a plurality of the directed acyclic graphs to determined similar interpretations of the multimedia content;
traversing the multimedia content according to the directed acyclic graph and the measured attributes assigned to the content; and
summarizing the multimedia content according to the directed acyclic graph and the measured attributes assigned to the content entities.
2. The method of claim 1 wherein the measured attributes include intensity attributes.
3. The method of claim 1 wherein the measured attributes include direction attributes.
4. The method of claim 1 wherein the measured attributes include spatial attributes and the order is spatial.
5. The method of claim 1 wherein the measured attributes include temporal attributes and the order is temporal.
6. The method of claim 1 wherein the measured attributes are arranged in an increasing rank order.
7. The method of claim 1 wherein the measured attributes are arranged in an decreasing rank order.
8. The method of claim 1 wherein the multimedia content is a three dimensional video sequence.
9. The method of claim 1 wherein nodes of the directed acyclic graphs represent the content entities and edges represent breaks in the segmentation, and the measured attributes are associated with the corresponding edges.
10. The method of claim 1 wherein at least one secondary content entity is associated with a particular content entity, and wherein the secondary content entity is selected during the traversing.
11. The method of claim 1 wherein a summary of the multimedia is a selected permutation of the content entities according to the associated ranks.