CONTEXTUAL INFORMATION OF VISUAL MEDIA

• US 20160140398A1
• Filed: 07/01/2015
• Published: 05/19/2016
• Est. Priority Date: 11/14/2014
• Status: Active Grant

• Current Assignee: Artax LLC (IPValuation Partners, LLC)
• Original Assignee: TeleCommunication Systems Inc (Comtech Telecommunications Corporation)
• Inventors: PITT, LANCE D., Hines, Gordon John
• Granted Patent: US 9,514,368 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: G06T 2207/20164 Salient point detection; Corner detection
• CPC Class Codes: G06T 2210/61 Scene description
• 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 20/582 of traffic signs
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names

Abstract

An analysis master control can be configured to derive contextual information of visual media that includes extracted information and extrapolated information. The analysis master control can receive the extracted information characterizing visual media from a recognizer. An information finder can be configured to query a plurality of information sources for information based on the extracted information. The analysis master control can also be configured to match information received from the information sources with the extracted information to form the extrapolated information that characterizes the visual media.

Claims

1. A non-transitory machine readable medium having machine executable instructions comprising:

an analysis master control configured to derive contextual information of visual media that includes extracted information and extrapolated information, wherein the analysis master control receives the extracted information characterizing visual media from a recognizer; and

an information finder configured to query a plurality of information sources for information based on the extracted information;

wherein the analysis master control is further configured to match information received from the information sources with the extracted information to form the extrapolated information that characterizes the visual media.

2. The medium of claim 1, wherein the visual media is a still frame.

3. The medium of claim 1, further comprising a presentation builder configured to generate a contextual document that includes the visual media and the contextual information.

4. The medium of claim 2, wherein the contextual document includes text and a hyperlink characterizing content of the visual media.

5. The medium of claim 1, wherein the recognizer is further configured to:

identify line segments in the visual media;

identify a shape in the visual media based on the line segments; and

identify text in the visual media base on at least one of the identified line segments and shape.

6. The medium of claim 5, wherein the recognizer is further configured to identify a corner between line segments in the visual media.

7. The medium of claim 5, wherein the recognizer is further configured to identify a street sign in the visual media based on the identified shape.

8. The medium of claim 6, wherein the recognizer is further configured to estimate a distance between objects in the visual media based on the identified street sign.

9. The medium of claim 6, wherein the recognizer is further configured to estimate a perspective of the visual media based on the identified street sign.

10. The medium of claim 1, wherein the information finder is further configured to query a search engine based on the extracted information.

11. The medium of claim 1, wherein the information finder is further configured to query a database based on the extracted information.

12. The medium of claim 1, wherein the contextual information includes text that characterizes a location corresponding to the visual media.

13. The medium of claim 1, wherein the contextual information includes text that characterizes a make of a vehicle in the visual media.

14. The medium of claim 1, wherein the contextual information includes text that characterizes a business name included in the visual media.

15. The medium of claim 1, wherein the contextual information includes text that characterizes a partial address included in the visual media.

16. A system comprising:

a media content information resolver (MCIR) configured to receive visual media from a media source;

an analysis master control configured to derive contextual information for the visual media that characterizes information extracted from the visual media and information extrapolated for the visual media based on the extracted information; and

a presentation builder configured to generate a contextual document for the MCIR based on the contextual information.

17. The system of claim 16, further comprising an information finder configured to query a plurality of information sources for information based on the extracted information to facilitate determination of the extrapolated information.

18. The system of claim 16, wherein the MCIR is further configured to output the contextual information for the visual media via a graphical user interface (GUI).

19. A method comprising:

recognizing shapes in visual media based on identified line segments;

extracting information from the visual media, wherein the extracted information includes text present in the visual media;

retrieving information from at least one of a search engine or a database based on the extracted information; and

determining extrapolated information based on the retrieved information, wherein contextual information for the visual media includes the extracted information and the extrapolated information.

20. The method of claim 19, further comprising generating a contextual document that includes the visual media and the contextual information for the visual media.