Using meta-descriptors to represent multimedia information

A method of representing a plurality of multimedia information, comprising:

• acquiring descriptors for the multimedia information;
• generating clusters of the descriptors;
• generating at least one meta-descriptors for the clusters;
• and
• attaching the at least one meta-descriptor to the multimedia information, including respectively attaching the meta-descriptors for the clusters to items of the multimedia information described by the descriptors in the clusters,
• wherein the meta-descriptor generating step comprises generating respective groups of data elements for each of the clusters indicating relevancy of the descriptors therein, and
• wherein at least some of the descriptors are representations of features of an item of multimedia information belonging to a category of multimedia content, the features comprising an ordered set of features including color, texture, shape and sketch, and the category of multimedia content being still image;
• and
• the meta-descriptor generating step comprises generating respective groups of weight values for each of the clusters indicating respective weights for the descriptors therein.

A method of representing a plurality of multimedia information, comprising:

• acquiring descriptors for the multimedia information;
• generating clusters of the descriptors;
• generating at least one meta-descriptors for the clusters;
• and
• attaching the at least one meta-descriptor to the multimedia information, including respectively attaching the meta-descriptors for the clusters to items of the multimedia information described by the descriptors in the clusters,
• wherein the meta-descriptor generating step comprises generating respective groups of data elements for each of the clusters indicating relevancy of the descriptors therein, and wherein:
• at least some of the descriptors are representations of features of an item of multimedia information belonging to a category of multimedia content, the features comprising an ordered set of features including color, texture, shape and sketch, and the category of multimedia content being still image;
• and
• the meta-descriptor generating step comprises generating respective character strings for each of the clusters identifying at least one relevant feature having a predetermined set of representation types, and at least one of the representation types from the predetermined set of representation types.

A method of representing a plurality of multimedia information, comprising:

• acquiring descriptors for the multimedia information;
• generating at least one meta-descriptors for the descriptors and a group of data elements indicating relevancy of the descriptors therein;
• and
• attaching the at least one meta-descriptor to the multimedia information wherein
• at least some of the descriptors are representations of features of an item of multimedia information belonging to a category of multimedia content, the features comprising an ordered set of features including color, texture, shape and sketch, and the category of multimedia content being still image;
• and
• the meta-descriptor generating step comprises generating binary vectors indicating relevancy of the descriptors therein.

A method of representing a plurality of multimedia information, comprising:

• acquiring descriptors for the multimedia information;
• generating at least one meta-descriptors for the descriptors and a group of data elements indicating relevancy of the descriptors therein;
• and
• attaching the at least one meta-descriptor to the multimedia information wherein
• at least some of the descriptors are representations of features of an item of multimedia information belonging to a category of multimedia content, the features comprising an ordered set of features including color, texture, shape and sketch, and the category of multimedia content being still image;
• and
• the meta-descriptor generating step comprises generating a group of weight values indicating weights for the descriptors therein.

A method of representing a plurality of multimedia information, comprising:

• acquiring descriptors for the multimedia information;
• generating at least one meta-descriptors for the descriptors and a group of data elements indicating relevancy of the descriptors therein;
• and
• attaching the at least one meta-descriptor to the multimedia information wherein
• at least some of the descriptors are representations of features of an item of multimedia information belonging to a category of multimedia content, the features comprising an ordered set of features including color, texture, shape and sketch, and the category of multimedia content being still image;
• and
• the meta-descriptor generating step comprises generating a character string identifying at least one relevant feature having a predetermined set of representation types, and at least one of the representation types from the predetermined set of representation types.