Video content claiming classifier

• US 8,706,675 B1
• Filed: 08/29/2011
• Issued: 04/22/2014
• Est. Priority Date: 08/29/2011
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: King, David G., Lin, Ruei-Sung, Samaniego, Clifford, Ross, David A., Frank, Alexander Joshua, Madani, Omid, Arai, Kenji
• Primary Examiner(s): CHANG, LI WU
• Application Number:
  US13/220,378
• Time in Patent Office: 967 Days
• Field of Search: None
• US Class Current: 706/52
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0241 Advertisements
• CPC Class Codes: G06Q 30/0251 Targeted advertisements

Abstract

A video hosting service comprising video classifiers that identify content sources of content included in videos uploaded to the video hosting service. Identifying the content source allows a content owner of the content source to claim ownership of videos that include content based on the content source. Usage policies associated with the content owners are applied to the uploaded videos that describe how the video hosting service is to treat the videos.

Claims

1. A computer-implemented method of identifying a usage policy of a video, the method comprising:

receiving a video comprising content captured from a perspective of a user observing an event that is represented by the content of the video;

extracting features from the video, the features comprising audiovisual features extracted from the content of the video;

applying a plurality of classifiers to the video, each classifier trained to identify a content source that is associated with the classifier;

identifying from among the plurality of classifiers, a classifier that generated a score exceeding a threshold value;

assigning a title of a content source associated with the identified classifier to the video;

identifying a usage policy for the video based on the title of the content source, the usage policy describing actions to apply to the video; and

associating the usage policy with the video.

2. The computer-implemented method of claim 1, wherein the content of the video comprises gameplay of a video game.

3. The computer-implemented method of claim 1, wherein the content of the video comprises footage of a live concert.

4. The computer-implemented method of claim 1, wherein the content of the video comprises footage of a live sporting event.

5. The computer-implemented method of claim 1, wherein the content of the video comprises animation created by a user.

6. The computer-implemented method of claim 1, wherein the features further comprise textual features extracted from the content of the video.

7. The computer-implemented method of claim 1, wherein applying the plurality of classifiers comprises, for each classifier:

generating a classification score indicating a measure of likelihood that the content of the video represents a content source associated with the classifier.

8. The computer-implemented method of claim 1, wherein the content source comprises content owned by a content owner.

9. The computer-implemented method of claim 1, wherein the usage policy comprises an action to monetize the video, wherein monetizing the video comprises providing advertisements in conjunction with the video when the video is served.

10. The computer-implemented method of claim 1, wherein the usage policy comprises an action to block the video, wherein blocking the video comprises preventing the video from being displayed to one or more client devices.

11. The computer-implemented method of claim 1, wherein the usage policy comprises an action to track the video, wherein tracking the video comprises monitoring usage interactions with the video to determine whether to monetize the video.

12. The computer-implemented method of claim 1, wherein applying the plurality of classifiers comprises:

applying a second plurality of classifiers to the video, each classifier in the second plurality of classifiers trained to identify a genre that is associated with the classifier; and

wherein each of the plurality of classifiers applied to the video are associated with the identified genre.

13. A computer-implemented method of training a plurality of video classifiers, the method for training each classifier comprising:

accessing a training set of videos associated with a video classifier that is untrained, the training set of videos comprising content that the video classifier is being trained to recognize;

deriving a set of features from the training set, the set of features comprising audiovisual features extracted from the content of the training set;

applying a plurality of trained genre classifiers to the training set of videos;

generating a plurality genre scores based on the application of the plurality of trained genre classifiers, each genre score indicating a likelihood that the content of the training set of videos is of a genre associated with a genre classifier that generated the genre score;

training the video classifier based on the derived set of features and the genre scores, the trained video classifier when applied to a video comprising content captured from a perspective of a user observing an event that is represented by the content of the video produces a classification score indicating a likelihood that content of the video represents the content from the training set of videos; and

storing the trained video classifier.

14. The computer-implemented method of claim 13, further comprising:

training the plurality of genre classifiers, each genre classifier trained with a set of videos including content of a genre associated with the genre classifier; and

storing the trained plurality of genre classifiers.

15. The computer-implemented method of claim 13, wherein the content associated with the video classifier comprises gameplay of a particular video game.

16. The computer-implemented method of claim 13, wherein the content associated with the video classifier comprises footage of a concert.

17. The computer-implemented method of claim 13, wherein the content associated with the video classifier comprises footage of a sporting event.

18. A computer-implemented method of identifying a content owner of a video, the method comprising:

accessing a training set of videos associated with a video classifier that is untrained, the training set of videos comprising content that the video classifier is being trained to recognize;

deriving a set of features from the training set, the set of features comprising audiovisual features extracted from the content of the set of videos;

applying a plurality of trained genre classifiers to the training set of videos;

generating a plurality genre scores based on the application of the plurality of trained genre classifiers, each genre score indicating a likelihood that the content of the training set of videos is of a genre associated with a genre classifier that generated the genre score;

training the video classifier based on the derived set of features and the genre scores, the trained video classifier when applied to videos produces classification scores indicating a likelihood that content of the videos represents the content from the training set, wherein each video comprises content captured from a perspective of a user observing an event that is represented by the content of the video;

storing the trained video classifier;

receiving a video;

extracting features from the video, the features comprising audiovisual features extracted from the content of the video;

applying a plurality of classifiers including the trained video classifier to the video, each classifier trained to identify a content source that is associated with the classifier;

classifying the content of the video based on the application of the plurality of classifiers;

identifying a usage policy for the video based on the classification, the usage policy describing actions to apply to the video; and

associating the usage policy with the video.

19. A computer program product comprising a non-transitory computer-readable storage medium containing executable computer program code for identifying a usage policy of a video, the code when executed are for:

receiving a video comprising content captured from a perspective of a user observing an event that is represented by the content of the video;

extracting features from the video, the features comprising audiovisual features extracted from the content of the video;

applying a plurality of classifiers to the video, each classifier trained to identify a content source that is associated with the classifier;

identifying from among the plurality of classifiers, a classifier that generated a score exceeding a threshold value;

assigning a title of a content source associated with the identified classifier to the video;

identifying a usage policy for the video based on the title of the content source, the usage policy describing actions to apply to the video; and

associating the usage policy with the video.

20. The computer program product of claim 19, wherein the content of the video comprises gameplay of a video game, footage of a live concert, a live sporting event, or animation created by a user.

21. The computer program product of claim 19, wherein the code when executed is further for:

generating a classification score indicating a measure of likelihood that the content of the video represents a content source associated with the classifier.

22. The computer program product of claim 19, wherein the usage policy comprises an action to monetize the video, wherein monetizing the video comprises providing advertisements in conjunction with the video when the video is served.

23. The computer program product of claim 19, wherein the usage policy comprises an action to block the video, wherein blocking the video comprises preventing the video from being displayed to one or more client devices.

24. The computer program product of claim 19, wherein the usage policy comprises an action to track the video, wherein tracking the video comprises monitoring usage interactions with the video to determine whether to monetize the video.

25. The computer program product of claim 19, wherein the code when executed is further for:

applying a second plurality of classifiers to the video, each classifier in the second plurality of classifiers trained to identify a genre that is associated with the classifier; and

wherein each of the plurality of classifiers applied to the video are associated with the identified genre.

26. A computer program product comprising a non-transitory computer-readable storage medium containing executable computer program code for training a plurality of video classifiers, the code when executed are for:

accessing a training set of videos associated with a video classifier that is untrained, the training set of videos comprising content that the video classifier is being trained to recognize;

deriving a set of features from the training set, the set of features comprising audiovisual features extracted from the content of the training set;

applying a plurality of trained genre classifiers to the training set of videos;

generating a plurality genre scores based on the application of the plurality of trained genre classifiers, each genre score indicating a likelihood that the content of the training set of videos is of a genre associated with a genre classifier that generated the genre score;

training the video classifier based on the derived set of features and the genre scores, the trained video classifier when applied to a video comprising content captured from a perspective of a user observing an event that is represented by the content of the video produces a classification score indicating a likelihood that content of the video represents the content from the training set of videos; and

storing the trained video classifier.

27. The computer program product of claim 26, wherein the code when executed is further for:

training the plurality of genre classifiers, each genre classifier trained with a set of videos including content of a genre associated with the genre classifier; and

storing the trained plurality of genre classifiers.

28. The computer program product of claim 26, wherein the content comprises gameplay of a video game, footage of a live concert, a live sporting event, or animation created by a user.

29. A computer system for identifying a usage policy of a video, the system comprising:

a computer processor;

a computer-readable storage medium comprising executable computer program code when executed by the computer processor is for;

receiving a video comprising content captured from a perspective of a user observing an event that is represented by the content of the video;

extracting features from the video, the features comprising audiovisual features extracted from the content of the video;

applying a plurality of classifiers to the video, each classifier trained to identify a content source that is associated with the classifier;

identifying from among the plurality of classifiers, a classifier that generated a score exceeding a threshold value;

assigning a title of a content source associated with the identified classifier to the video;

identifying a usage policy for the video based on the title of the content source, the usage policy describing actions to apply to the video; and

associating the usage policy with the video.

30. A computer system for training a plurality of video classifiers, the system comprising:

a computer processor;

a computer-readable storage medium comprising executable computer program code when executed by the computer processor is for;

accessing a training set of videos associated with a video classifier that is untrained, the training set of videos comprising content that the video classifier is being trained to recognize;

deriving a set of features from the training set, the set of features comprising audiovisual features extracted from the content of the training set;

applying a plurality of genre classifiers to the training set of videos;

generating a plurality genre scores based on the application of the plurality of genre classifiers, each genre score indicating a likelihood that the content of the training set of videos is of a genre associated with the genre classifier that generated the genre score;

training the video classifier based on the derived set of features and the genre scores, the trained video classifier when applied to a video comprising content captured from a perspective of a user observing an event that is represented by the content of the video produces a classification score indicating a likelihood that content of the video represents the content from the training set of videos; and

storing the trained video classifier.