IP Library Granted Patent US 8,332,412
Granted Patent B2
US 8,332,412 · App. 12/603,356 · Granted Dec 11, 2012

Method and apparatus for staged content analysis

Assignee: AT&T Intellectual Property I, LP
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 8,332,412
App. No.
12/603,356
Granted
Dec 11, 2012
Kind
B2
Abstract

A system that incorporates teachings of the present disclosure may include, for example network device having a controller to receive multiple streams of content for portions of a multimedia work (MMW), perform a high level analysis for features in each of the streams for the MMW, perform a specialized analysis on the portion having a detected general feature to generate a content analysis output, correlate the content analysis output with other content analysis of the MMW, and output a weighted content description based on the correlation function. Other embodiments are disclosed.

Claims (32)

1. A method comprising:

receiving, by a system including a processor, a content stream of a multimedia work;

performing, by the system, a coarse feature analysis of the content stream at a multiple frame level;

identifying, by the system, one or more portions of the content stream having one or more potential occurrences of one or more objectionable features of a plurality of objectionable features according to the coarse feature analysis of the content stream;

responsive to identifying the one or more portions of the content stream, performing, by the system, one or more feature specific analyses on the one or more portions of the content stream at a frame by frame level;

identifying, by the system, one or more actual occurrences of the one or more objectionable features according to the one or more feature specific analyses of the one or more portions of the content stream; and

correlating, by the system, the identified one or more actual occurrences of the one or more objectionable features with additional information associated with the multimedia work to generate a weighted content description of the multimedia work.

2. The method of claim 1 , comprising applying a policy engine with rules to generate the weighted content description of the multimedia work.

3. The method of claim 1 , comprising generating statistics associated with the weighted content description of the multimedia work.

4. The method of claim 1 , comprising obtaining the additional information from a metadata source associated with the multimedia work.

5. The method of claim 1 , comprising correlating the identified one or more actual occurrences of the one or more objectionable features with additional information associated with the multimedia work and an external database to generate a second weighted content description of the multimedia work.

6. The method of claim 1 , wherein the feature specific analyses comprise at least one among an obscenity analysis, an audio violence analysis, a gunshot analysis or a combination thereof.

7. The method of claim 1 , comprising filtering out redundant content in the content stream prior to performing the coarse feature analysis.

8. The method of claim 1 , wherein the one or more feature specific analyses comprise determining temporal content redundancies in the one or more portions of the content stream.

9. The method of claim 1 , comprising receiving the content stream of the multimedia work from an Internet Protocol television network.

10. A network device, comprising a memory coupled to a controller, wherein the memory comprises computer instructions, which when executed by the controller, cause the controller to perform operations comprising:

receiving a content stream of a multimedia work;

identifying one or more portions of the content stream having one or more potential occurrences of one or more objectionable features from a plurality of objectionable features according to a coarse feature analysis of the content stream at a multiple frame level;

responsive to identifying the one or more portions of the content stream, identifying one or more actual occurrences of the one or more objectionable features according to one or more feature specific analyses of the one or more portions at a frame by frame level; and

correlating the identified one or more actual occurrences of the one or more objectionable features with additional information associated with the multimedia work to generate a weighted content description of the multimedia work.

11. The network device of claim 10 , wherein the one or more feature specific analyses comprise an obscenity analyzer, an audio violence analyzer, or a gunshot analyzer.

12. The network device of claim 10 , wherein the computer instructions cause the controller to apply a policy engine with rules to the weighted content description of the multimedia work.

13. The network device of claim 10 , wherein the network device is a set top box in an interactive television network.

14. The network device of claim 10 , wherein the computer instructions cause the controller to obtain the additional information from at least one of a metadata source associated with the multimedia work or an external database.

15. The network device of claim 10 , wherein the computer instructions cause the controller to filter out redundant content in the content stream.

16. A non-transitory computer-readable storage medium, comprising computer instructions, which when executed by a processor of a network device cause the processor to perform operations comprising:

identifying one or more portions of a received content stream having one or more potential occurrences of one or more objectionable features from a plurality of objectionable features according to a coarse feature analysis of the received content stream at a multiple frame level;

responsive to identifying that one or more portions of the received content stream, identifying one or more actual occurrences of the one or more objectionable features according to one or more feature specific analyses of the one or more portions at a frame by frame level; and

correlating the identified one or more actual occurrences of the one or more objectionable features with additional information associated with the multimedia work to generate a weighted content description of the multimedia work.

17. The non-transitory computer-readable storage medium of claim 16 , wherein the network device comprises computer instructions to cause the processor to output the weighted content description of the multimedia work.

18. The non-transitory computer-readable storage medium of claim 16 , wherein the one or more feature specific analyses comprise an obscenity analyzer, an audio violence analyzer, or a gunshot analyzer.

19. The non-transitory computer-readable storage medium of claim 16 , wherein the network device comprises computer instructions to cause the processor to filter out redundant content.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 11, 2011
From: BASSO, ANDREA; DE LOS REYES, GUSTAVO
To: AT&T INTELLECTUAL PROPERTY I, L.P.
Reel/Frame 025937/0613 →
Continuity (1)
Related Publication 20110093473A1 · Apr 21, 2011