Image composition evaluation

• US 8,132,096 B1
• Filed: 03/26/2007
• Issued: 03/06/2012
• Est. Priority Date: 03/26/2007
• Status: Expired due to Fees

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Widdowson, Simon, O'Brien-Strain, Eamonn
• Primary Examiner(s): Huynh, Thu
• Application Number:
  US11/728,654
• Time in Patent Office: 1,807 Days
• Field of Search: 715/202, 715/243, 715/716, 715/723, 715/255, 358/3.06, 358/3.01, 345/626, 382/209, 382/282
• US Class Current: 715/243
• CPC Class Codes: G06T 11/60 Editing figures and text; Combining figures or text
• CPC Class Codes: H04N 1/387 Composing, repositioning or otherwise geometrically modifying originals

Abstract

In a method of evaluating compositional quality of a layout orientation of an image, image saliency data is received for an image to be presented in a viewing region. A layout orientation for presenting the image in the viewing region is received. A composition template embodying an image composition rule is received. The image saliency data and the composition template are utilized to determine a weighted occlusion cost for the layout orientation. The weighted occlusion cost is utilized to determine a composition score indicating quality of the layout orientation with respect to the image composition rule.

Claims

1. A computer implemented method of evaluating compositional quality of a layout orientation of an image, said method comprising:

receiving image saliency data for an image to be presented in a viewing region;

receiving a layout orientation for presenting said image in said viewing region;

receiving a composition template embodying an image composition rule, wherein said composition template is a grayscale map and wherein said composition template is not derived from said image;

utilizing said image saliency data, said composition template, and summing values of a plurality of pixels of a weighted saliency distribution to determine a weighted occlusion cost for viewing said image using said layout orientation; and

utilizing said weighted occlusion cost to determine a composition score indicating quality of said layout orientation with respect to said image composition rule.

2. The method as recited in claim 1, wherein receiving image saliency data for an image to be presented in a viewing region comprises:

receiving said image saliency data represented in the form of a grayscale map of said image.

3. The method as recited in claim 1, wherein said receiving a composition template embodying an image composition rule comprises:

receiving a saliency map, said saliency map designating an area of visual interest in a viewing region as expressed by said image composition rule.

4. The method as recited in claim 1, wherein said receiving a composition template embodying an image composition rule comprises:

receiving a saliency map wherein a saliency value assigned to an area within said saliency map is weighted to express an importance of visibility of said area according to said image composition rule.

5. The method as recited in claim 1, wherein said receiving a composition template embodying an image composition rule comprises:

receiving said composition template from a library of composition templates.

6. The method as recited in claim 1, wherein said utilizing said image saliency data and said composition template to determine a weighted occlusion cost for said layout orientation comprises:

utilizing said image saliency data and said layout orientation to determine a layout saliency distribution;

utilizing said composition template and said layout saliency distribution to determine said weighted saliency distribution of said image layout; and

determining said weighted occlusion cost from said weighted saliency distribution.

7. The method as recited in claim 1, wherein said utilizing said weighted occlusion cost to determine a composition score indicating quality of said layout orientation with respect to said image composition rule comprises:

determining a ratio of said weighted occlusion cost to a summation of visible saliency in a layout saliency distribution of said layout orientation.

8. A computer-readable medium having computer-executable instructions for performing steps of a method for determining a layout orientation for presenting an image within a viewing region, said instructions comprising:

receiving image saliency data for an image to be displayed within a viewing region;

receiving a composition template embodying an image composition rule, wherein said composition template is a grayscale map and wherein said composition template is not derived from said image;

determining a plurality of layout orientations for presenting said image in said viewing region;

utilizing said image saliency data, said composition template, and summing values of a plurality of pixels of a weighted saliency distribution to determine a weighted occlusion cost for viewing said image using a layout orientation of said plurality of layout orientations;

utilizing said weighted occlusion cost to determine a composition score indicating quality of said layout orientation with respect to said image composition rule; and

determining a selected layout orientation for presenting said image within said viewing region by comparing composition scores for said plurality of layout orientations.

9. The computer-readable medium of claim 8, wherein said method further comprises computer-executable instructions for:

cropping said image according to dimensions of said selected layout orientation for presenting said image within said viewing region.

10. The computer-readable medium of claim 8, wherein said computer executable instructions for receiving image saliency data for an image to be presented in a viewing region comprise computer-executable instructions for:

receiving said image saliency data represented in the form of a grayscale map of said image.

11. The computer-readable medium of claim 8, wherein said computer executable instructions for receiving a composition template embodying an image composition rule comprise computer-executable instructions for:

receiving a saliency map, said saliency map designating an area of visual interest in said viewing region as expressed by said image composition rule.

12. The computer-readable medium of claim 8, wherein said computer executable instructions for receiving a composition template embodying an image composition rule comprise computer-executable instructions for:

receiving a saliency map wherein a saliency value assigned to an area within said saliency map is weighted to express an importance of visibility of said area according to said image composition rule.

13. The computer-readable medium of claim 8, wherein said computer executable instructions for utilizing said image saliency data and said composition template to determine a weighted occlusion cost for a layout orientation of said plurality of layout orientations comprise computer-executable instructions for:

utilizing a scaled composition template to determine said weighted saliency distribution for a portion of a layout saliency distribution; and

determining said weighted occlusion cost from said weighted saliency distribution.

14. The computer-readable medium of claim 8, wherein said computer executable instructions for utilizing said weighted occlusion cost to determine a composition score indicating quality of said layout orientation with respect to said image composition rule comprise computer-executable instructions for:

determining a ratio of said weighted occlusion cost to a summation of visible saliency in a layout saliency distribution of said layout orientation.

15. A method for capturing an image composition rule in a composition template, said method comprising:

presenting a viewing region to a user, wherein said viewing region is presented to said user in a uniform color;

receiving a user designation of an area of visual interest in said viewing region, said area of visual interest representing a user specified image composition rule regarding positioning of a salient image feature with respect to said viewing region;

generating an updated viewing region from said viewing region, said updated viewing region indicating said area of interest with a color other than said uniform color;

saving said updated viewing region as a composition template which embodies said image composition rule, wherein said composition template is a grayscale map and wherein said composition template is not derived from an image;

utilizing an image saliency data, said composition template, and summing values of a plurality of pixels of a weighted saliency distribution to determine a weighted occlusion cost for viewing said image using a layout orientation; and

utilizing said weighted occlusion cost to determine a composition score indicating quality of said layout orientation with respect to said image composition rule.

16. The method as recited in claim 15, wherein said presenting a viewing region to a user comprises:

presenting said viewing region as an image comprised of a uniform grayscale color.

17. The method as recited in claim 15, wherein said receiving a user designation of an area of visual interest in said viewing region comprises:

receiving a user designation of a plurality of areas of visual interest in said viewing region.

18. The method as recited in claim 15, wherein said receiving a user designation of an area of visual interest in said viewing region comprises:

receiving a user designation of a plurality of areas of varying levels of visual interest in said viewing region.

19. The method as recited in claim 18, wherein said generating an updated viewing region from said viewing region comprises:

utilizing a plurality of grayscale colors to indicate said plurality of areas of varying levels of visual interest, wherein each level of said varying levels of visual interest is represented by a unique grayscale color.

20. The method as recited in claim 15, wherein said saving said updated viewing region as a composition template which embodies said image composition rule comprises:

saving said composition template as a saliency map for use in evaluating compositional quality of a digital image.