Computation of intrinsic perceptual saliency in visual environments, and applications

• US 20060215922A1
• Filed: 05/08/2006
• Published: 09/28/2006
• Est. Priority Date: 03/08/2001
• Status: Active Grant

• Current Assignee: California Institute of Technology
• Original Assignee: California Institute of Technology
• Inventors: Itti, Laurent, Koch, Christof
• Granted Patent: US 8,098,886 B2
• Time in Patent Office: Days
• US Class Current: 382/248
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06V 10/454 Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]
• CPC Class Codes: G06V 10/462 Salient features, e.g. scale invariant feature transforms [SIFT]

Abstract

Detection of image salience in a visual display of an image. The image is analyzed at multiple spatial scales and over multiple feature channels to determine the likely salience of different portions of the image. One application for the system is in an advertising context. The detection may be improved by second order statistics, e.g. mean and the standard deviations of different image portions relative to other portions. Different edges may be considered as being extended edges by looking at the edges over multiple spatial scales. One set of feature channels can be optimized for use in moving images, and can detect motion or flicker. The images can be obtained over multiple spectral ranges the user can be instructed about how to maximize the saliency. This can be applied to automatically evaluate and optimize sales or advertisement displays.

Claims

1-14. -14. (canceled)

15. A method, comprising:

using a computing device to analyze an image representation to predict which parts in said image may be perceived as visually conspicuous by a human observer, and

using said predicted visually conspicuous parts to determine the effectiveness of said image in displaying its contents.

16. A method as in claim 15 wherein said parts may attract the visual attention of said observer.

17. A method as in claim 15, wherein said parts may attract the eye gaze of said human observer.

18. A method as in claim 15, wherein said image representation may be generated by a computing device.

19. A method as in claim 15, wherein said image representation may be a digital capture of some portion of the real world.

20. A method as in claim 15, wherein said perceived is by an animal observer.

21. A method as in claim 15, wherein said perceived is by a robotic device.

22. A method as in claim 15, wherein said analyzing comprises analyzing pixels of said image using mean pixel values.

23. A method as in claim 15, wherein said analyzing comprises analyzing pixels on said image using higher order statistical variations.

24. A method as in claim 15, wherein said image representation includes a single image at a single time.

25. A method as in claim 15, wherein said image representation includes a sequence of images over time.

26. A method as in claim 15, wherein said using comprises evaluating an effectiveness of said image in an advertising context.

27. A method as in claim 15, wherein said using comprises evaluating a display showing one or more items for sale.

28. A method as in claim 15, wherein said analyzing comprises obtaining information about the image in at least a first spectral range and a second spectral range that is a different spectral range than said first spectral range;

and correlating said information about the image in said first spectral range to information in said second spectral range.

29. A method as in claim 15, further comprising using said analyzing to optimize a display of visual information.