Implicit user interest marks in media content

• US 9,378,286 B2
• Filed: 03/14/2008
• Issued: 06/28/2016
• Est. Priority Date: 03/14/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Sloo, David, Barrett, Peter
• Primary Examiner(s): APONTE, FRANCISCO JAVIER
• Application Number:
  US12/049,142
• Publication Number:
  US 20110173194A1
• Time in Patent Office: 3,028 Days
• Field of Search: 707/705, 707/736, 707741-751, 706 12- 14, 706 45- 61, 715/716
• US Class Current: 1/1
• CPC Class Codes: G06F 16/745 the internal structure of a single video sequence
• CPC Class Codes: G06F 16/7867 using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

Automatically marking locations of user interest in media content is disclosed. Data that describes user interaction with media content is analyzed to determine locations of potential interest. For each of the locations of potential interest, the data is analyzed to determine whether user interest satisfies a criterion for establishing a mark for the location. Marks are determined for those locations that satisfy the criterion. A mark includes a description for a given location and information that allows the given location to be automatically presented to a user upon selection of the description. In another aspect, extrinsic data that is separate from a piece of media content is analyzed to determine whether moments of high user interest might exist in the piece of media content. Then, the piece of media content is scanned to find locations in the media content that correspond to the moments.

Claims

1. A computer implemented method of providing marks associated with media content, said method comprising:

automatically searching extrinsic information over a network for a summary description of an event, the summary description stating times in the event at which respective key moments occurred in the event;

automatically analyzing said summary description of the event in order to determine whether moments of high user interest might exist in a piece of media content that contains the event;

accessing audio/video content of the event, wherein the summary description of the event is derived independently from user interaction with audio/video content of the event;

determining locations in the accessed audio/video content that correspond to each of the moments of high user interest, including scanning the accessed audio/video content to find locations in the accessed audio/video content that correspond to the times in the event at which respective key moments occurred in the event, wherein the times in the event are independent of elapsed time in the accessed audio/video content;

determining an identifier of each of the locations, said identifier for a given location allows the given location to be determined by an electronic device;

associating a unique description with each of the identifiers; and

providing the unique descriptions and the identifiers of the locations to at least one user electronic device to enable selective viewing of portions of the audio/video content that correspond to the moments of high user interest.

2. The computer implemented method of claim 1, wherein the searching extrinsic information for the summary description of the event includes:

accessing the summary description from a first content provider that is distinct from a second content provider that provides the accessed audio/video content.

3. The computer implemented method of claim 2, wherein the searching extrinsic information for the summary description of the event includes:

automatically performing a Web search to locate the summary description of the event.

4. The computer implemented method of claim 1, further comprising:

analyzing the accessed audio/video content to determine the unique description that is associated with each of the identifiers.

5. The computer implemented method of claim 1, further comprising:

analyzing the summary description of the event to determine the unique description that is associated with each of the identifiers.

6. The computer implemented method of claim 1, wherein the determining locations in the accessed audio/video content that correspond to each of the moments of high user interest that might exist in the piece of media content includes using feature extraction to locate the moments of high user interest that might exist in the piece of media content.

7. The method of claim 1, wherein automatically analyzing said summary description of the event in order to determine whether moments of high user interest might exist in a piece of media content that contains the event comprises:

automatically reading a sports box score to determine at what time in the event scores or other key moments occurred.

8. A system comprising:

processor readable storage that stores instructions; and

a processor in communication with the processor readable storage, wherein the instructions which when executed on the processor cause the processor to;

automatically search extrinsic information for a summary description of an event;

automatically analyze said summary description of the event in order to determine key moments in the event, the summary description stating times in the event at which the respective key moments occurred in the event;

access audio/video content of the event, wherein the audio/video content is independent from the extrinsic information;

determine a location in the audio/video content that corresponds to each of the key moments in the event, including causing the processor to scan the audio/video content to find locations in the audio/video content that correspond to the times in the event at which the respective key moments occurred, the times in the event being independent of elapsed time in the audio/video content;

determine an identifier of each of the locations, said identifier for a given location allows the given location to be determined by an electronic device;

associate a unique description with each of the identifiers; and

provide the unique descriptions and the identifiers of the locations to at least one user electronic device to enable selective viewing of portions of the audio/video content that correspond to the key moments in the event.

9. The system of claim 8, wherein the times in the summary description are game times of a sporting event.

10. The system of claim 8, wherein the summary description of the event is a sports box score.

11. The system of claim 8, wherein the instructions that cause the processor to automatically search extrinsic information for a summary description of the event cause the processor to automatically perform a Web search to locate the summary description of the event.

12. The system of claim 8, wherein the summary description of the event is derived independently of both user interaction with audio/video content of the event and user commentary about the event.

13. A media analyzer comprising:

an interface connected to a network; and

a processing unit connected to the interface, wherein the processing unit is configured to;

perform a search over the network for a summary of an event;

analyze the summary of the event in order to determine key moments of expected user interest in the event, said summary stating times in the event at which the key moments occurred in the event;

access audio/video content of the event over the network, wherein the summary is derived independently from user interaction with audio/video content of the event;

scan the audio/video content to find locations that correspond to the key moments of expected user interest in the event, including the processing unit configured to scan the audio/video content to find locations in the audio/video content that correspond to the times in the event at which the key moments occurred, wherein the times in the event are independent of elapsed time in the audio/video content;

establish a media mark for each of the locations that correspond to the key moments of expected user interest in the event, wherein the media mark for a given key moment of expected user interest has a unique description of the key moment of expected user interest and an identifier of the location of the key moment of expected user interest; and

provide the media mark of each of the locations over the network to a user electronic device.

14. The media analyzer of claim 13, wherein the times in the summary are game times in a sporting event.

15. The media analyzer of claim 13, wherein the summary description of the event is derived independently of both user interaction with audio/video content of the event and user feedback about the audio/video content.