Method for enriching a multimedia content, and corresponding device

• US 9,740,695 B2
• Filed: 07/11/2014
• Issued: 08/22/2017
• Est. Priority Date: 07/12/2013
• Status: Expired due to Fees

• Current Assignee: Magnolia Licensing LLC (IP Edge LLC)
• Original Assignee: Thomson Licensing (Vantiva SA)
• Inventors: Lanagan, James, Guegan, Marie, Schmouker, Philippe
• Primary Examiner(s): Al Hashemi, Sana
• Application Number:
  US14/329,018
• Publication Number:
  US 20150019555A1
• Time in Patent Office: 1,138 Days
• Field of Search: 707607, 707609, 707687, 707705, 707790, 707813, 707821, 707953, 707964
• 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 16/4393 Multimedia presentations, e.g. slide shows, multimedia albums
• CPC Class Codes: G06F 16/45 Clustering; Classification
• CPC Class Codes: G06F 16/48 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually

Abstract

According to the invention, the method comprises the following steps of:

• • clustering said plurality of creators to deliver clusters comprising at least two creators, called relevant creators;
  • assigning, to at least some relevant creators of the cluster comprising the first creator, called first cluster, a score of relevance determined from at least one criterion of similarity between said relevant creators and said first creator;
  • ranking relevant creators of said first cluster according to their scores of relevance;
  • selecting, among a plurality of multimedia contents, at least one second multimedia content related to said first multimedia content and delivered by a ranked relevant creator of said first cluster;
  • enriching said first multimedia content based on said second multimedia content.

Claims

1. Method for enriching a first multimedia content created by a first creator, among a plurality of multimedia content creators within an online community, comprising, at a microprocessor:

forming clusters, at a clustering circuit of said microprocessor, from said plurality of multimedia content creators, each cluster comprising at least two relevant creators from said plurality of multimedia content creators;

assigning, at an assigning circuit of said microprocessor, to a first cluster including the first creator, a score of relevance to each relevant creator within the first cluster, said score of relevance being determined from at least one criterion of similarity between each respective relevant creator and said first creator;

ranking, at a ranking circuit of said microprocessor, said relevant creators of said first cluster according to scores of relevance assigned to each relevant creator of said first cluster;

selecting, at a selecting circuit of said microprocessor, among a plurality of multimedia contents from the ranked relevant creators, at least one second multimedia content related to said first multimedia content for enriching the first multimedia content;

retrieving said at least one second multimedia content selected from the ranked relevant creators; and

enriching, at an enriching circuit of said microprocessor, said first multimedia content based on said at least one second multimedia content.

4. Method according to claim 1, further comprising extracting, from said second multimedia content, at least one element related to said first multimedia content.

5. Method according to claim 1, wherein said clustering implements a K-means algorithm based on a set of identified features.

6. Method according to claim 1, wherein said selected second multimedia content is created by the ranked relevant creator of said first cluster having a highest score of relevance among assigned scores of relevance.

7. Method according to claim 1, further comprising selecting several second multimedia contents related to said first multimedia content and receiving said selected several second multimedia contents from ranked relevant creators of said first cluster having a score of relevance at least equal to a threshold of relevance.

8. Method according to claim 1, wherein said criterion of similarity is a distance criterion.

9. Method according to claim 1, wherein said first multimedia content is a textual content.

16. Method according to claim 4, wherein the extracting relies on natural language processing.

2. Device for enriching a first multimedia content created by a first creator, among a plurality of multimedia content creators within an online community, wherein the device comprises a microprocessor comprising:

a clustering circuit configured to form clusters from said plurality of multimedia content creators, each cluster comprising at least two relevant creators from said plurality of multimedia content creators;

an assigning circuit configured to assign, to a first cluster including the first creator, a score of relevance to each relevant creator within the first cluster, said score of relevance being determined from at least one criterion of similarity between each respective relevant creator and said first creator;

a ranking circuit configured to rank said relevant creators of the first cluster according to scores of relevance assigned to each relevant creator of said first cluster;

a selecting circuit configured to select, among a plurality of multimedia contents from the ranked relevant creators, at least one second multimedia content related to the first multimedia content for enriching the first multimedia content and retrieving said at least one second multimedia content selected from the ranked relevant creators of the first cluster;

an enriching circuit configured to enrich the first multimedia content based on said at least one second multimedia content.

10. Device according to claim 2, further comprising an extracting circuit adapted to extract, from said second multimedia content, at least one element related to said first multimedia content.

11. The device according to claim 2, wherein said clustering circuit forms clusters by implementing a K-means algorithm based on a set of identified features.

12. The device according to claim 2, wherein said second multimedia content is created by the ranked relevant creators of said first cluster having a highest score of relevance among assigned scores of relevance.

13. The device according to claim 2, wherein said selecting circuit selects several second multimedia contents related to said first multimedia content and receives said selected several second multimedia contents from ranked relevant creators of said first cluster having a score of relevance at least equal to a threshold of relevance.

14. The device according to claim 2, wherein said at least one criterion of similarity is a distance criterion.

15. The device according to claim 2, wherein said first multimedia content is a textual content.

17. The device according to claim 10, wherein said extracting circuit extracts using natural language processing.

3. Non-transitory computer-readable medium comprising a computer program product recorded thereon and capable of being run by a processor, including program code instructions for

forming clusters, at a clustering circuit of said microprocessor, from said plurality of multimedia content creators, each cluster comprising at least two relevant creators from said plurality of multimedia content creators;

assigning, at an assigning circuit of said microprocessor, to a first cluster including the first creator, a score of relevance to each relevant creator within the first cluster, said score of relevance being determined from at least one criterion of similarity between each respective relevant creator and said first creator;

ranking, at a ranking circuit of said microprocessor, said relevant creators of said first cluster according to scores of relevance assigned to each relevant creator of said first cluster;

selecting, at a selecting circuit of said microprocessor, among a plurality of multimedia contents from the ranked relevant creators, at least one second multimedia content related to said first multimedia content for enriching the first multimedia content;

retrieving said at least one second multimedia content selected from the ranked relevant creators; and

enriching, at an enriching circuit of said microprocessor, said first multimedia content based on said at least one second multimedia content.