Network for describing multimedia information

• US 7,146,349 B2
• Filed: 11/01/2001
• Issued: 12/05/2006
• Est. Priority Date: 11/06/2000
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Benitez-Jimenez, Ana Belen, Smith, John R., Li, Chung-Sheng
• Primary Examiner(s): Knight, Anthony
• Assistant Examiner(s): Williams, Jr., Ronald E.
• Application Number:
  US10/002,998
• Publication Number:
  US 20020107827A1
• Time in Patent Office: 1,860 Days
• Field of Search: 706/46, 706/55, 707/3, 382/190, 382/195, 717/143, 704/500, 704/9, 704/1
• US Class Current: 706/46
• CPC Class Codes: G06F 16/40 of multimedia data, e.g. slideshows comprising image and additional audio data retrieval of still image data G06F16/50; retrieval of audio data G06F16/60; retrieval of video data G06F16/70
• CPC Class Codes: G06F 16/48 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually

Abstract

A method and apparatus for encoding knowledge using a multimedia network. A multimedia network represents semantic concepts and their relations using multimedia content. A multimedia network associates words and multimedia content with the semantic concepts in order to illustrate and exemplify the semantic concepts as well as describe lexical, semantic, and perceptual relations. The multimedia network can be searched, browsed, or summarized for purposes of information discovery. The multimedia network can also be used for personalizing multimedia content or for querying a multimedia information repository by expanding a query to include related concepts encoded in a Multimedia network.

Claims

1. A method implemented by at least one computer for encoding knowledge, comprising the steps of:

forming a network having nodes that represent semantic concepts;

associating one or more words with one or more of the nodes;

associating multimedia content with one or more of the nodes;

representing relationships between the nodes as arcs between associated words and arcs between associated multimedia content;

receiving a user query for at least one semantic concept;

recursively searching the network for matching multimedia content within the network related to the user query; and

creating a new multimedia presentation from the matching multimedia content within the network.

2. The method of claim 1, further comprising:

creating lexical relations between semantic concepts on the basis of one or more of;

word forms and word meaning of associated words.

3. The method of claim 1, wherein relationships between semantic concepts and between associated content are based at least in part on audio and/or visual feature descriptor values.

4. The method of claim 3, further comprising:

extracting feature descriptors from multimedia content; and

computing similarity measures between descriptor values.

5. The method of claim 1, wherein the media network knowledge is represented using the ISO MPEG-7 Description Definition Language.

6. The method of claim 1, wherein the relationships between the nodes are based, at least in part, on the features of the multimedia content.

7. The method of claim 1, wherein the relationships between the nodes denote similarity of semantic concepts.

8. The method of claim 1, further comprising displaying the new multimedia presentation on a monitor.

9. A method implemented by at least one computer for searching an encoded media network knowledge representation that comprises a network having nodes that represent semantic concepts, one or more words and multimedia associated with the one or more nodes, and wherein relationships between the nodes are represented as arcs between associated words and arcs between associated multimedia content, the method comprising the steps of:

accepting a query;

matching the query to the words and multimedia content related to the concepts encoded in the media network knowledge representation;

navigating the relationship arcs of the concepts associated with matching words and multimedia content;

retrieving related concepts, words, and multimedia content from the matched nodes or related nodes;

creating a new multimedia presentation from the matching related concepts, words, and multimedia content.

10. The method of claim 9, further comprising:

forming a query comprised of words; and

matching the query words to the words encoded in the media network knowledge representation.

11. The method of claim 9, further comprising:

forming a query comprised of multimedia content; and

matching the query content to the multimedia content encoded in the media network knowledge representation.

12. The method of claim 9, further comprising:

forming a query comprised of audio and/or visual feature descriptor values, wherein the feature descriptor values denote proximity to the semantic concepts of the nodes; and

matching the query descriptor values to the descriptor values of the content encoded in the media network knowledge representation.

13. A computer-implemented method for browsing an encoded media network knowledge representation that comprises a network having nodes that represent semantic concepts, one or more words and multimedia associated with the one or more nodes, and wherein relationships between the nodes are represented as arcs between associated words and arcs between associated multimedia content, the method comprising the steps of:

displaying one or more concept nodes and associated words and/or multimedia content;

providing means for allowing a user to select related concepts for viewing;

receiving a user query for at least one semantic concept;

recursively searching the network for matching multimedia content within the network related to the user query; and

creating a new multimedia presentation from the matching multimedia content within the network.

14. The method of claim 13, further comprising:

providing means for allowing the user to select concept nodes and associated words and/or multimedia content for display on the basis of specific types or values of relations to a particular concept node or associated word or multimedia content.

15. A method implemented by at least one computer for summarizing an encoded media network knowledge representation that comprises a network having nodes that represent semantic concepts, one or more words and multimedia associated with the one or more nodes, and wherein relationships between the nodes are represented as arcs between associated words and arcs between associated multimedia content, the method comprising the steps of:

extracting a subset of nodes, relations, and words and/or multimedia content from an encoded media network knowledge representation;

receiving a user query for at least one semantic concept;

recursively searching the network for matching multimedia content within the network related to the user query; and

creating a new multimedia presentation from the matching multimedia content within the network.

16. The method of claim 15, further comprising:

consolidating together concept nodes, relations, words, and/or multimedia content.

17. A method implemented by at least one computer for updating an encoded media network knowledge representation that comprises a network having nodes that represent semantic concepts, one or more words and multimedia associated with the one or more nodes, and wherein relationships between the nodes are represented as arcs between associated words and arcs between associated multimedia content, the method comprising the steps of:

adding, deleting or modifying concepts, relations, or associated words, multimedia content, or descriptors in the encoded media network knowledge representation;

receiving a user query for at least one semantic concept;

recursively searching the network for matching multimedia content within the network related to the user query; and

creating a new multimedia presentation from the matching multimedia content within the network.

18. A method implemented by at least one computer for querying a multimedia information repository associated with an encoded media network knowledge representation that comprises an encoded network having nodes that represent semantic concepts, one or more words and multimedia associated with the one or more nodes, and wherein relationships between the nodes are represented as arcs between associated words and arcs between associated multimedia content, the method comprising the steps of:

searching the encoded media network knowledge representation;

retrieving words, content, and/or descriptors from the media network knowledge representation;

searching the information repository using the retrieved words, content, and/or descriptors; and

creating a new multimedia presentation from the retrieved words, content, and/or descriptors within the network.

19. A method implemented by at least one computer for personalizing multimedia information in a system comprising an encoded media network knowledge representation that includes an encoded network having nodes that represent semantic concepts, one or more words and multimedia associated with the one or more nodes, and wherein relationships between the nodes are represented as arcs between associated words and arcs between associated multimedia content, the method comprising the steps of:

describing the multimedia information using words or descriptors;

describing user preferences using words, multimedia content, and/or descriptors;

matching the user preferences with the descriptions of the multimedia information; and

extracting, retrieving, and/or summarizing the matched multimedia items; and

creating a new multimedia presentation from the matched multimedia items.

20. A computer program product in a computer readable medium for use for encoding knowledge, the computer program product comprising:

first instructions for forming a network having logical nodes that represent semantic concepts;

second instructions for associating one or more words with one or more of the nodes;

third instructions for associating multimedia content with one or more of the nodes;

fourth instructions for representing relationships between the nodes as arcs between associated words and arcs between associated multimedia content;

fifth instructions for receiving a user query for at least one semantic concept;

sixth instructions for recursively searching the network for matching multimedia content within the network related to the user query; and

seventh instructions for creating a new multimedia presentation from the matching multimedia content within the network.