Personalized Event Notification Using Real-Time Video Analysis

• US 20090249387A1
• Filed: 03/31/2008
• Published: 10/01/2009
• Est. Priority Date: 03/31/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: El-Saban, Motaz, Magdy, Walid
• Granted Patent: US 8,881,191 B2
• Time in Patent Office: Days
• US Class Current: 725/32
• CPC Class Codes: G06F 16/7335 Graphical querying, e.g. query-by-region, query-by-sketch, query-by-trajectory, GUIs for designating a person/face/object as a query predicate end-user interface involving hot spots associated with the video H04N21/4725; end-user interface for selecting a Region of Interest H04N21/4728
• CPC Class Codes: G06F 16/7834 using audio features
• CPC Class Codes: H04H 60/48 for recognising items expressed in broadcast information
• CPC Class Codes: H04H 60/65 for using the result on users' side

Abstract

A system and methods for personalized event notification using real-time video analysis are described. In an embodiment, a live video feed is automatically monitored and characteristics of the video feed are analyzed against criteria which relate to an event which has been identified by a user. The criteria relate to characteristics of the image or the audio in the video feed. If the criteria are satisfied, the user is notified that the event has occurred and this notification may be achieved through messages on the video display screen (e.g. on the television screen) or other messaging techniques (e.g. SMS or email). In an example, the criteria may define a particular logo and the user is notified if the analysis detects the logo in the video feed.

Claims

1. A method of personalized event notification comprising:

receiving a media feed;

analyzing the media feed in real-time to detect a user specified event; and

on detection of the user specified event, sending a notification message.

2. A method according to claim 1, wherein analyzing the media feed in real-time to detect a user specified event comprises:

performing pattern matching to determine if a portion of the media feed matches a target pattern associated with the user specified event.

3. A method according to claim 2, wherein the target pattern comprises at least one of an audio pattern and an image pattern.

4. A method according to claim 2, wherein the media feed comprises a video feed and target pattern comprises a measure of change between frames in the portion of the video feed.

5. A method according to claim 2, wherein performing pattern matching to determine if a portion of the media feed matches a target pattern associated with the user specified event comprises:

accessing the target pattern;

comparing the target pattern and the portion of the media feed and saving a score indicating a level of correspondence between the target pattern and the portion; and

detecting the user specified event when the score exceeds a threshold.

6. A method according to claim 5, wherein comparing the target pattern and the portion of the media feed and saving a score indicating a level of correspondence between the target pattern and the portion comprises:

(i) searching a first frame in the portion of the media feed for the target pattern and saving a score indicating a level of correspondence between the target pattern and the first frame;

(ii) if the score exceeds a first threshold, incrementing a match score; and

repeating steps (i) and (ii) for each frame in the portion of the media feed, and wherein detecting the user specified event when the score exceeds a threshold comprises;

detecting the user specified event when the match score exceeds a second threshold.

7. A method according to claim 1, wherein analyzing the media feed in real-time to detect a user specified event comprises:

classifying a portion of the media feed; and

determining if a classification of the portion of the media feed matches a classification associated with the user specified event.

8. A method according to claim 1, wherein sending a notification message comprises:

sending a notification message to a user.

9. A method according to claim 1, wherein sending a notification message comprises:

displaying a notification message on a video display device.

10. A method according to claim 1, further comprising:

receiving a user input identifying the media feed;

receiving data from a user identifying the user specified event and a notification method associated with the user specified event; and

storing the data.

11. A method according to claim 10, wherein receiving data from a user identifying the user specified event and a notification method associated with the user specified event comprises:

displaying a video clip;

enabling a user to freeze the video clip on a frame and select a portion of said frame; and

saving the portion of said frame as a target pattern.

12. A method according to claim 1, wherein the media feed comprises at least one broadcast television channel.

13. An automated event detection system comprising:

an event detection module arranged to receive and analyze a media feed in real-time to detect a user specified event in the media feed; and

a notification module arranged to send a notification on detection of the user specified event.

14. An automated event detection system according to claim 13, further comprising a data store comprising metadata associated with the user specified event, and wherein the event detection module is further arranged to access the metadata and to use the data to detect the user specified event in the media feed.

15. An automated event detection system according to claim 13, wherein the event detection module comprises a feature extraction module arranged to extract a set of features from a portion of the media feed and a feature analysis module arranged to detect the user specified event based on the set of features.

16. An automated event detection system according to claim 13, wherein the event detection module comprises a classifier.

17. An automated event detection system according to claim 13, further comprising a communication interface and wherein the notification module is arranged to send a notification message to a user associated with the user specified event via the communication interface.

18. An automated event detection system according to claim 13, further comprising a media recording module, wherein the notification module is arranged to send the notification to the media recording module to trigger recording of the media feed.

19. One or more tangible device-readable media with device-executable instructions for performing steps comprising:

analyzing a portion of a media feed in real-time to determine, based on metadata associated with a user specified event, if the portion of the media feed includes the user specified event; and

if the portion of the media feed includes the user specified event, sending a notification to a user associated with the user specified event.

20. One or more tangible device-readable media according to claim 19, wherein analyzing a portion of a media feed in real-time to determine, based on metadata associated with a user specified event, if the portion of the media feed includes the user specified event comprises:

capturing a frame from the media feed;

matching the captured frame and a target pattern associated with the user specified event to produce a matching score; and

determining if the matching score exceeds a threshold.