Computerized systems and methods for generating models for identifying thumbnail images to promote videos

• US 9,449,231 B2
• Filed: 11/12/2014
• Issued: 09/20/2016
• Est. Priority Date: 11/13/2013
• Status: Active Grant

• Current Assignee: Yahoo AD Tech LLC
• Original Assignee: AOL Advertising Incorporated (Apollo Global Management, Inc.)
• Inventors: Batiz, Jaime, Luenberger, Robert, Karlsson, Niklas
• Primary Examiner(s): NGUYEN, HUY THANH
• Application Number:
  US14/539,729
• Publication Number:
  US 20150131967A1
• Time in Patent Office: 678 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 18/41 Interactive pattern learning with a human teacher
• CPC Class Codes: G06V 10/7788 the supervisor being a human, e.g. interactive learning with a human teacher
• CPC Class Codes: G06V 20/46 Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
• CPC Class Codes: G06V 20/49 Segmenting video sequences, i.e. computational techniques such as parsing or cutting the sequence, low-level clustering or determining units such as shots or scenes

Abstract

Systems, methods, and computer-readable media are provided for generating and using models to select thumbnail images for videos. In one exemplary method, the method comprises extracting at least one thumbnail from a video and determining at least one feature present in the at least one thumbnail. The method further comprises sending the extracted at least one thumbnail to one of an editor or at least one viewer. The method further comprises receiving feedback information related to the at least one thumbnail. The method further comprises generating a model and storing the generated model for use in identifying thumbnails for other videos. The model can be generated based on the determined at least one feature and the received feedback information.

Claims

1. A computerized method comprising the following operations performed by at least one processor:

extracting at least one thumbnail from a video;

determining at least one feature present in the extracted at least one thumbnail;

sending the extracted at least one thumbnail to one of an editor or at least one viewer;

receiving, in response to sending the extracted at least one thumbnail to the editor or the at least one viewer, feedback information related to the at least one thumbnail, the feedback information comprising at least one of (i) response data associated with the at least one viewer or (ii) data identifying thumbnails selected by the editor;

generating a model based on the determined at least one feature, the feedback information, and a classification of the at least one thumbnail into one or more groups; and

storing the generated model for use in identifying thumbnails for other videos.

2. The method of claim 1, wherein the determined at least one feature comprises at least one of blurriness of the at least one thumbnail or video, colors present in the at least one thumbnail or video, presence or absence of items in the at least one thumbnail or video, or metadata associated with the video.

3. The method of claim 1, further comprising:

receiving a second video; and

determining, based on the generated model, thumbnails to extract from the second video.

4. The method of claim 1, further comprising determining at least one classification associated with the video, wherein generating the model further comprises generating the model based on the at least one classification associated with the video.

5. The method of claim 1, wherein extracting thumbnails further comprises:

comparing pixel data of a first thumbnail to pixel data of a second thumbnail; and

based on the comparison, utilizing at least one of the first thumbnail and second thumbnail as the extracted at least one thumbnail.

6. The method of claim 1, wherein generating the model comprises utilizing one or more weak classifiers with a boosting algorithm.

19. The method of claim 1, further comprising:

determining at least one classification associated with the video; and

identifying at least one stored model based on the at least one classification associated with the video.

7. A tangible computer-readable storage medium, comprising instructions that, when executed by at least one processor, cause the at least one processor to perform the following operations:

extracting at least one thumbnail from a video;

determining at least one feature present in the extracted at least one thumbnail;

sending the extracted at least one thumbnail to one of an editor or at least one viewer;

receiving, in response to sending the extracted at least one thumbnail to the editor or the at least one viewer, feedback information related to the at least one thumbnail, the feedback information comprising at least one of (i) response data associated with the at least one viewer or (ii) data identifying thumbnails selected by the editor;

generating a model based on the determined at least one feature, the feedback information, and a classification of the at least one thumbnail into one or more groups; and

storing the generated model for use in identifying thumbnails for other videos.

8. The medium of claim 7, wherein the determined at least one feature comprises at least one of blurriness of the at least one thumbnail or video, colors present in the at least one thumbnail or video, presence or absence of items in the at least one thumbnail or video, or metadata associated with the video.

9. The medium of claim 7, wherein the instructions further cause the at least one processor to perform operations comprising:

receiving a second video; and

determining, based on the generated model, thumbnails to extract from the second video.

10. The medium of claim 7, wherein the instructions further cause the at least one processor to perform operations comprising determining at least one classification associated with the video, wherein generating the model further comprises generating the model based on the at least one classification associated with the video.

11. The medium of claim 7, wherein extracting thumbnails further comprises:

comparing pixel data of a first thumbnail to pixel data of a second thumbnail; and

based on the comparison, utilizing at least one of the first thumbnail and second thumbnail as the extracted at least one thumbnail.

12. The medium of claim 7, wherein generating the model comprises utilizing one or more weak classifiers with a boosting algorithm.

20. The medium of claim 7, wherein the instructions further cause the at least one processor to perform operations comprising:

determining at least one classification associated with the video; and

identifying at least one stored model based on the at least one classification associated with the video.

13. A computerized system, comprising:

at least one processor; and

a storage medium comprising instructions that, when executed by the at least one processor, configure the at least one processor to perform the following operations;

extracting at least one thumbnail from a video;

determining at least one feature present in the extracted at least one thumbnail;

sending the extracted at least one thumbnail to one of an editor or at least one viewer;

receiving, in response to sending the extracted at least one thumbnail to the editor or the at least one viewer, feedback information related to the at least one thumbnail, the feedback information comprising at least one of (i) response data associated with the at least one viewer or (ii) data identifying thumbnails selected by the editor;

generating a model based on the determined at least one feature, the feedback information, and a classification of the at least one thumbnail into one or more groups; and

storing the generated model for use in identifying thumbnails for other videos.

14. The system of claim 13, wherein the determined at least one feature comprises at least one of blurriness of the at least one thumbnail or video, colors present in the at least one thumbnail or video, presence or absence of items in the at least one thumbnail or video, or metadata associated with the video.

15. The system of claim 13, wherein the instructions further cause the at least one processor to perform operations comprising:

receiving a second video; and

determining, based on the generated model, thumbnails to extract from the second video.

16. The system of claim 13, further wherein the instructions further cause the at least one processor to perform operations comprising determining at least one classification associated with the video, wherein generating the model further comprises generating the model based on the at least one classification associated with the video.

17. The system of claim 13, wherein extracting thumbnails further comprises:

comparing pixel data of a first thumbnail to pixel data of a second thumbnail; and

based on the comparison, utilizing at least one of the first thumbnail and second thumbnail as the extracted at least one thumbnail.

21. The system of claim 13, wherein the instructions further cause the at least one processor to perform operations comprising:

determining at least one classification associated with the video; and

identifying at least one stored model based on the at least one classification associated with the video.

18. A computerized method comprising the following operations performed by at least one processor:

determining, based on a first video, a model for extracting one or more thumbnails from the first video;

extracting one or more thumbnails from the first video;

selecting at least one of the one or more extracted thumbnails;

sending the selected at least one thumbnail to one of an editor or at least one viewer;

receiving, in response to sending the selected at least one thumbnail to the editor or the at least one viewer, feedback information related to the at least one thumbnail, the feedback information comprising at least one of (i) response data associated with the at least one viewer or (ii) data identifying thumbnails selected by the editor; and

updating the determined model based on the feedback information.