Granted Patent B2

US 9,064,154 · App. 13/533,429 · Granted Jun 23, 2015

Abstract

Claims (87)

1. A computer-implemented method for identifying associated video content, comprising:

decomposing first and second elements of video content into corresponding first and second segments;

computing measures of similarity between candidate pairs of the first and second video segments;

determining, using at least one processor, that the similarity measures corresponding to a subset of the candidate pairs satisfy at least one association rule;

for a plurality of the candidate pairs of the subset, determining, using the processor, that corresponding ones of the first segments are disposed adjacently within the first video content element, and that corresponding ones of the second segments are disposed adjacently within the second video content element;

merging, using the at least one processor, (i) the corresponding first segments to generate a merged first segment and (ii) the corresponding second segments to generate a merged second segment, the merged first and second segments establishing a merged pair of first and second segments; and

generating a value indicative of an association between the merged pair of first and second segments,

wherein computing measures of similarity comprises:

extracting a first auditory sample from at least one of the first video segments and a second auditory sample from at least one of the second video segments, the first and second auditory samples comprising pluralities of frames;

generating representations of a power spectrum associated with the frames of the first and second auditory samples; and

computing the similarity measure for the at least one first and second video segments based on the generated power spectrum representations.

2. The method of claim 1 , further comprising obtaining first metadata associated with the first element of video content and second metadata associated with the second element of video content.

3. The method of claim 2 , wherein:

the method further comprises determining whether the first and second elements of video content satisfy the at least one association rule, based on the first and second metadata; and

wherein the decomposing comprises decomposing the first and second elements of video content into the corresponding first and second segments, when the first and second elements of video content satisfy the at least one association rule.

4. The method of claim 2 , further comprising:

obtaining information associated with a characteristic of the first element of video content or the second element of video content;

incorporating the obtained information within the metadata of the associated first and second elements of video content.

5. The method of claim 2 , wherein the computing comprises:

extracting at least one of auditory samples or visual samples from the first and second video segments;

calculating fingerprints corresponding to the extracted samples; and

computing the similarity measure for the first and second video segments based on the fingerprints.

6. The method of claim 1 , wherein:

the association rule identifies a threshold value for the similarity measure; and

determining whether the similarity measure satisfies the association rule comprises determining whether the similarity value exceeds the threshold value.

7. The method of claim 1 , wherein:

the association rule identifies specifies a characteristic of video content; and

determining whether the similarity measure satisfies the association rule comprises determining whether the first and second elements of video content are associated with the specified characteristic.

8. The method of claim 7 , wherein the characteristic comprises at least one of a predetermined duration, a video quality, a video type, a performer associated with the video content, or a source of the video content.

9. The method of claim 1 , wherein the first and second elements of video content comprise video clips.

10. The method of claim 1 , further comprising generating video graph data indicative of a link between the first and second elements of video content, based on at least the value indicative of the association between the first and second elements of video content.

11. The method of claim 1 , further comprising:

receiving information associated with a third element of video content;

obtaining, in response to the received information, data associated with at least one video graph;

identifying one or more fourth elements of video content that are similar to the third element of video content, the identification being based on the obtained video graph data, the video graph data comprising information indicative of the similarity between the third and fourth elements of video content; and

identifying at least a subset of the third and fourth elements of video content for delivery to a user.

12. An apparatus, comprising:

a storage device; and

at least one processor coupled to the storage device, wherein the storage device stores a program for controlling the at least one processor, and wherein the at least one processor, being operative with the program, is configured to:

decompose first and second elements of video content into corresponding first and second segments;

compute measures of similarity between candidate pairs of the first and second video segments;

determine that the similarity measures corresponding to a subset of the candidate pairs satisfy at least one association rule;

for a plurality of candidate pairs of the subset, determine that corresponding ones of the first segments are disposed adjacently within the first video content element, and that corresponding ones of the second segments are disposed adjacently within the second video content element;

merge (i) the corresponding first segments to generate a merged first segment and (ii) the corresponding second segments to generate a merged second segment, the merged first and second segments establishing a merged pair of first and second segments; and

generate a value indicative of an association between the merged pair of first and second segments,

the processor being further configured to:

extract a first auditory sample from at least one of the first video segments and a second auditory sample from at least one of the second video segments, the first and second auditory samples comprising pluralities of frames:

generate representations of a power spectrum associated with the frames of the first and second auditory samples; and

compute the similarity measure for the at least one first and second video segments based on the generated power spectrum representations.

13. The apparatus of claim 12 , wherein the processor is further configured to:

obtain first metadata associated with the first element of video content and second metadata associated with the second element of video content;

determine whether the first and second elements of video content satisfy the at least one association rule, based on the first and second metadata; and

decompose the first and second elements of video content into the corresponding first and second segments, when the first and second elements of video content satisfy the at least one association rule.

14. The apparatus of claim 12 , wherein the processor is further configured to:

obtain information associated with a characteristic of the first element of video content or the second element of video content;

incorporate the obtained information within the metadata of the associated first and second elements of video content.

15. The apparatus of claim 12 , wherein the processor is further configured to:

extract at least one of auditory samples or visual samples from the first and second video segments;

calculate fingerprints corresponding to the extracted samples; and

compute the similarity measure for the first and second video segments based on the fingerprints.

16. The apparatus of claim 12 , wherein:

the association rule identifies a threshold value for the similarity measure; and

the processor is further configured to determine whether the similarity measure satisfies the association rule comprises determining whether the similarity value exceeds the threshold value.

17. The apparatus of claim 12 , wherein:

the association rule identifies specifies a characteristic of video content; and

the processor is further configured to determine whether the similarity measure satisfies the association rule comprises determining whether the first and second elements of video content are associated with the specified characteristic.

18. The apparatus of claim 12 , wherein the first and second elements of video content comprise video clips.

19. The apparatus of claim 12 , wherein the processor is further configured to generate video graph data indicative of a link between the first and second elements of video content, based on at least the value indicative of the association between the first and second elements of video content.

20. The apparatus of claim 12 , wherein the processor is further configured to:

receive information associated with a third element of video content;

obtain, in response to the received information, data associated with at least one video graph;

identify one or more fourth elements of video content that are similar to the third element of video content, the identification being based on the obtained video graph data, the video graph data comprising information indicative of the similarity between the third and fourth elements of video content; and

identify at least a subset of the third and fourth elements of video content for delivery to a user.

21. A tangible, non-transitory computer-readable medium storing instructions that, when executed by at least one processor, perform a method identifying associated video content, the method comprising the steps of:

decomposing first and second elements of video content into corresponding first and second segments;

computing measures of similarity between candidate pairs of the first and second video segments;

determining that the similarity measures corresponding to a subset of the candidate pairs satisfy at least one association rule;

for a plurality of candidate pairs of the subset, determining that corresponding ones of the first segments are disposed adjacently within the first video content element, and that corresponding ones of the second segments are disposed adjacently within the second video content element;

merging the corresponding first segments to generate a merged first segment, and the corresponding second segments to generate a merged second segment, the merged first and second segments establishing a merged pair of first and second segments; and

generating a value indicative of an association between the merged pair of first and second segments,

wherein computing measures of similarity comprises:

extracting a first auditory sample from at least one of the first video segments and a second auditory sample from at least one of the second video segments, the first and second auditory samples comprising pluralities of frames;

generating representations of a power spectrum associated with the frames of the first and second auditory samples; and

computing the similarity measure for the at least one first and second video segments based on the generated power spectrum representations.

22. The method of claim 1 , further comprising:

obtaining information identifying a prior consumption of at least one of audio or video content by a user; and

identifying at least one of the first or second video content based on the prior consumption information.

Assignments (6)