Characterizing dynamic regions of digital media data

• US 20070104369A1
• Filed: 11/06/2006
• Published: 05/10/2007
• Est. Priority Date: 11/04/2005
• Status: Active Grant

• Current Assignee: EyeTracking, Inc.
• Original Assignee: EyeTracking, Inc.
• Inventors: Weatherhead, James
• Granted Patent: US 8,155,446 B2
• Time in Patent Office: Days
• US Class Current: 382/181
• CPC Class Codes: G06Q 30/0242 Determining effectiveness of advertisements
• CPC Class Codes: H04H 60/45 for identifying users
• CPC Class Codes: H04H 60/48 for recognising items expressed in broadcast information
• CPC Class Codes: H04H 60/59 of video recognising characters or patterns in general G06F18/00, G06V20/00
• CPC Class Codes: H04H 60/66 for using the result on distributors' side

Abstract

A media analysis tool is provided for defining dynamic regions of a digital media segment. The dynamic regions may contain at least part of a visible feature of the segment. Correlation of the defined regions with external data quantifying attention of a subject viewing the segment to locations on the screen provides measures of interest level and attention to visible features in the segment. The dynamic regions may be defined in only some of the frames of a segment. The dynamic region may be interpolated or extrapolated for frames in which it is not explicitly defined.

Claims

1. A computer-implemented method comprising:

defining a dynamic region in one or more frames of a digital media segment, the dynamic region containing at least part of a visible feature in the one or more frames;

analyzing a relationship between an external data set that is collected from one or more subjects viewing the digital media segment and the dynamic region; and

reporting a metric indicative of a subject interest level in the visible feature.

2. A computer-implemented method as in claim 1, wherein the defining the dynamic screen region comprises using an image recognition algorithm to automatically produce a mathematical definition of a screen region containing at least part of the visible feature.

3. A computer-implemented method as in claim 1, wherein the digital media segment is incorporated into a digital media presentation in a stochastically determined order.

4. A computer-implemented method as in claim 1, further comprising:

defining a second dynamic region in the digital media segment;

analyzing a relationship between the external data and the second dynamic region in the digital media segment; and

reporting a second metric indicative of a second subject interest level in the second dynamic region.

5. A computer-implemented method as in claim 1, further comprising:

defining a second dynamic region in the digital media segment; and

recording a metric related to intersection of the dynamic region with the second dynamic region.

6. A computer-implemented method as in claim 1, wherein the external data set comprises eye-tracking data for the one or more subjects.

7. A computer-implemented method as in claim 1, wherein the external data set comprises one or more of cognitive activity, cognitive engagement, and emotional response measurements.

8. A computer-implemented method as in claim 1, further comprising:

analyzing a relationship between a second external data set comprising one or more of cognitive activity, cognitive engagement, and emotional response measurements and the one or more dynamic region in the digital media segment; and

reporting a second metric indicative of one or more of a level of subject cognitive activity, a level of subject cognitive engagement, and a level of subject emotional response stimulated by the visible feature.

9. A computer-implemented method as in claim 1, further comprising generating a predicted definition of the dynamic region for an additional frame in which the dynamic region is not defined.

10. A computer-implemented method as in claim 9, wherein the generating comprises interpolating the predicted definition based on the on or more frames.

11. A computer-implemented method comprising

displaying a digital media segment comprising a plurality of frames to a subject;

recording a data set comprising screen locations gazed at by the subject while viewing the digital media segment;

selecting a display frame from the plurality of frames; and

plotting the data set in conjunction with the display frame to produce a snapshot comprising screen locations gazed at by the subject while viewing the digital media segment.

12. A computer-implemented method as in claim 12, further comprising:

defining a dynamic region in the digital media segment, the dynamic region being non-static in the plurality of frames; and

correlating the data set to the dynamic region such that the snapshot reflects a measure of subject attention to the dynamic region throughout the plurality of frames.

13. A computer-implemented method as in claim 12, further comprising:

recording a second data set comprising one or more of subject cognitive activity, subject cognitive engagement, and subject emotional response while viewing the digital media segment; and

plotting the second data set in conjunction with the snapshot.

14. A computer-implemented method as in claim 12, wherein the snapshot comprises a still image of the display frame overlaid with a series of connected points that show the screen locations gazed at by the subject while viewing the plurality of frames.

15. A computer-implemented method as in claim 12, wherein the snapshot comprises a still image of the display frame overlaid with an intensity plot that indicates an amount of time that the subject'"'"'s gaze fell within one or more regions on the display frame.

16. A computer-implemented method as in claim 12, further comprising:

aggregating additional data recorded while displaying the digital media segment to one or more additional subjects into the data set; and

plotting the aggregated data in conjunction with the snapshot.

17. A computer-implemented method comprising:

identifying a visible feature that is displayed in a digital media segment;

defining a first screen region that contains at least part of the visible feature in a first frame of the digital media segment;

defining a second screen region that contains at least part of the visible feature in a second frame of the digital media segment; and

predicting a definition of a third screen region that contains at least part of the visible feature in a third frame based on the definitions of the first screen region and the second screen region.

18. A computer-implemented method as in claim 17, wherein the defining of the first screen region comprises specifying a first location of a vertex point of a polygon in the first frame, and the defining of the second screen region comprises specifying a second location the vertex point in the second frame.

19. A computer-implemented method as in claim 18, wherein the predicting comprises interpolating a third location of the vertex point in the third frame.

20. A computer-implemented method as in claim 17, wherein the third frame appears between the first and the second frames.

21. A computer-implemented method as in claim 17, wherein the defining the first screen region comprises using a image recognition algorithm to automatically produce a mathematical definition of the first screen region that contains at least part of the visible feature.

22. A computer-implemented method as in claim 17, further comprising:

displaying the digital media segment to a subject;

recording a metric pertaining to a location of subject attention in the third frame; and

comparing the location of subject attention with the predicted definition of the third region to estimate subject interest in the visible feature.

23. A computer-implemented method comprising:

defining a plurality of segments in a digital media presentation;

presenting the digital media presentation to one or more subjects;

recording external data related to reactions of the subjects to the segments; and

correlating the external data to the plurality of segments to determine subject interest in each segment of the plurality of segments.

24. A computer-implemented method as in claim 23, wherein the external data are one or more of subject cognitive activity, subject cognitive engagement, and subject emotional response.