Keyword-tagging of scenes of interest within video content

• US 9,008,489 B2
• Filed: 02/19/2013
• Issued: 04/14/2015
• Est. Priority Date: 02/17/2012
• Status: Expired due to Fees

• Current Assignee: KDDI Corporation
• Original Assignee: KDDI Corporation
• Inventors: Hoashi, Keiichiro, Ono, Chihiro, Nakazawa, Masami
• Primary Examiner(s): DANG, HUNG Q
• Application Number:
  US13/770,272
• Publication Number:
  US 20130216203A1
• Time in Patent Office: 784 Days
• Field of Search: 386239-248
• US Class Current: 386/239
• CPC Class Codes: H04N 21/252 Processing of multiple end-users' preferences to derive collaborative data
• CPC Class Codes: H04N 21/41407 embedded in a portable device, e.g. video client on a mobile phone, PDA, laptop constructional details of equipment or arrangements specially adapted for portable computer application G06F1/1626; arrangements specially adapted for mobile receivers in broadcast systems H04H20/57
• CPC Class Codes: H04N 21/4312 involving specific graphical features, e.g. screen layout, special fonts or colors, blinking icons, highlights or animations
• CPC Class Codes: H04N 21/4316 for displaying supplemental content in a region of the screen, e.g. an advertisement in a separate window
• CPC Class Codes: H04N 21/47217 for controlling playback functions for recorded or on-demand content, e.g. using progress bars, mode or play-point indicators or bookmarks specific graphical features in visual interfaces H04N21/4312
• CPC Class Codes: H04N 21/4788 communicating with other users, e.g. chatting arrangements for providing for computer conferences, e.g. chat rooms, to substation in data switching networks H04L12/1813; distributed application using peer-to-peer [P2P] networks H04L67/104
• CPC Class Codes: H04N 21/488 Data services, e.g. news ticker systems specially adapted for using meteorological information in broadcast systems H04H60/71
• CPC Class Codes: H04N 21/8133 specifically related to the content, e.g. biography of the actors in a movie, detailed information about an article seen in a video program
• CPC Class Codes: H04N 21/8456 by decomposing the content in the time domain, e.g. in time segments
• CPC Class Codes: H04N 21/8547 involving timestamps for synchronizing content
• CPC Class Codes: H04N 21/8549 Creating video summaries, e.g. movie trailer retrieval in video databases by using presentations in form of a video summary G06F16/739
• CPC Class Codes: H04N 9/79 Processing of colour television signals in connection with recording

Abstract

Extraction of scenes of interest from video content is disclosed, which includes: collecting comments from a comment server, which contain keywords associated with the video content; segmenting the video content into scenes; measuring the number of ones of the comments which were posted for each scene, as a post count; detecting ones of the scenes, each of which has the post count equal to or larger than a predetermined number, as a plurality of scenes of interest; extracting, per each scene-of-interest, at least one of the keywords which is contained in the comments with a number equal to or larger than a pre-selected number, as a comment keyword; and tagging each scene of interest with the comment keyword, per each scene-of-interest.

Claims

1. A method of extracting scenes of interest from video content, the method comprising:

collecting a plurality of comments with posting timestamps indicating times at which the comments were posted, from a comment server, the comments containing keywords associated with the video content;

segmenting the video content into a plurality of scenes;

measuring a number of ones of the plurality of comments which were posted during a time zone of each scene, as a post count;

calculating an average μ

post count per scene and a standard deviation σ

of the post counts among each of the scenes;

detecting ones of the plurality of scenes having a post count equal to or larger than a sum of the average μ and

the standard deviation σ

as a plurality of scenes of interest; and

for each of the scenes of interest;

extracting at least one of the keywords which is contained in the plurality of comments with a number equal to or larger than a pre-selected number, as a comment keyword; and

tagging the scene of interest with the comment keyword.

2. The method according to claim 1, further comprising generating abbreviated content representative of an abbreviated version of the video content, by combining at least two of the scenes of interest.

3. The method according to claim 1, wherein the comment-keyword extraction operation for each of the scenes of interest comprises extracting candidate words from the plurality of comments, and extracting distinctive words as the comment keyword from the candidate words based on word frequencies.

4. The method according to claim 3,

wherein the comment-keyword extraction operation for each of the scenes of interest comprises:

extracting “

persons'"'"' names”

from the distinctive words; and

extracting an important word having a word frequency equal to or larger than a predetermined value, from the distinctive words; and

the comment-tagging operation for each of the scenes of interest comprises tagging each scene-of-interest with the comment keyword in the form of the part-of-speech of the “

person'"'"'s name” and

the important word.

5. The method according to claim 1, further comprising:

generating a post-count graph representative of a time-varying number of the posts, such that the post-count graph is tagged with the comment keywords at selected points in time; and

transmitting the post-count graph to a viewer'"'"'s terminal equipment allowing the viewer to view the video content.

6. The method according to claim 5, further comprising transmitting to the terminal equipment ones of the scenes of interest which are tagged with the requested keyword.

7. The method according to claim 5, further comprising transmitting to the terminal equipment ones of the plurality of comments which correspond to ones of the scenes of interest which are tagged with the requested keyword received from the terminal equipment.

8. The method according to claim 5, wherein the post-count graph generation operation comprises:

calculating a ratio of a number of users who posted the plurality of comments, to a total number of the plurality of comments, for ones of the scenes of interest which are tagged with the comment keywords on the post-count graph; and

displaying the comment keywords with which the scenes of interest are tagged, such that each comment keyword is visually emphasized at an increasing emphasis level as the ratio of the associated one of the tagged scenes of interest increases.

9. The method according to claim 5, wherein the post-count graph generation operation comprises:

calculating a ratio of a number of ones of the plurality comments which have quotation, to a total number of the plurality of comments, for ones of the scenes of interest which are tagged with the comment keywords on the post-count graph; and

displaying the comment keywords with which the scenes of interest are tagged, such that each comment keyword is visually emphasized at an increasing emphasis level as the ratio of the associated one of the tagged scenes of interest increases.

10. The method according to claim 5, wherein the video content includes at least one of TV broadcast content and content from a video sharing website, and the comment server includes a mini blog (mini Web log) server.

11. The method according to claim 10, wherein the mini blog server is a Twitter (Registered Trademark) server, and the comment collection operation comprises collecting the plurality of comments containing hash tags based on the video content, from the Twitter server.

12. The method according to claim 5, further comprising receiving a requested keyword from the terminal equipment, wherein the post-count graph generation operation comprises tagging the post-count graph with ones of the comment keywords which contain the requested keyword.

13. The method according to claim 12, wherein the post-count graph generation operation comprises generating a post-count graph representative of numbers of ones of the plurality of comments which contain the requested keyword.

14. A non-transitory storage medium having stored a computer program, when executed by a computer, for implementing the method according to claim 1.

15. A server of extracting scenes of interest from video content, comprising:

a collector that collects a plurality of comments with posting timestamps indicating times at which the comments were posted, from a comment server, the comments containing keywords associated with the video content;

a post counter that segments the video content into a plurality of scenes, and measures a number of ones of the plurality of comments which were posted during a time zone of each scene, as a post count;

a scene-of-interest detector that calculates an average μ

post count per scene and a standard deviation σ

of the post counts among each of the scenes, and detects ones of the plurality of scenes having a post count equal to or larger than a sum of the average μ and

the standard deviation σ

as a plurality of scenes of interest;

a comment-keyword extractor that extracts, for each of the scenes of interest, at least one of the keywords which is contained in the plurality of comments with a number equal to or larger than a pre-selected number, as a comment keyword;

a comment tagger that tags each scene of interest with the comment keyword; and

a scene-of-interest delivery unit that delivers the tagged scenes of interest, upon reception of a request from a viewer'"'"'s terminal equipment allowing the viewer to view the video content.

16. Terminal equipment in connection with a network, comprising:

a video-content receiver that receives video content to be published;

a collector collects a plurality of comments with posting timestamps indicating times at which the comments were posted, from a comment server, the comments containing keywords associated with the video content;

a post counter that segments the video content into a plurality of scenes, and measures a number of ones of the plurality of comments which were posted during a time zone of each scene, as a post count;

a scene-of-interest detector that calculates an average μ

post count per scene and a standard deviation σ

of the post counts among each of the scenes, and detects ones of the plurality of scenes having a post count equal to or larger than a stun of the average μ and

the standard deviation σ

, as a plurality of scenes of interest;

a comment-keyword extractor that extracts at least one of the keywords which is contained in the plurality of comments with a number equal to or larger than a pre-selected number, as a comment keyword; and

a comment tagger that tags each scene of interest with the comment keyword, for each of the scenes of interest.

17. A method of extracting scenes of interest from video content, the method comprising:

collecting a plurality of comments with posting timestamps indicating times at which the comments were posted, from a comment server, the comments containing keywords associated with the video content;

segmenting the video content into a plurality of scenes;

measuring a number of ones of the plurality of comments which were posted during a time zone of each scene, as a post count;

detecting ones of the plurality of scenes, each of which has a post count equal to or larger than a predetermined number, as a plurality of scenes of interest;

for each of the scenes of interest;

extracting at least one of the keywords which is contained in the plurality of comments with a number equal to or larger than a pre-selected number, as a comment keyword; and

tagging the scene of interest with the comment keyword;

generating a post-count graph representative of a time-varying number of the posts, such that the post-count graph is tagged with the comment keywords at selected points in time; and

transmitting the post-count graph to a viewer'"'"'s terminal equipment allowing the viewer to view the video content.

18. A non-transitory storage medium having stored a computer program, when executed by a computer, for implementing the method according to claim 17.