Image recomposition from face detection and facial features

1. A computing system comprising:

a memory configured to store a first digital image; and

a processing system configured to:

identify two or more individual regions in the first digital image that each include a face;

pad each of the two or more individual regions to produce two or more individual padded regions;

combine at least two of the two or more individual padded regions to form two or more combined padded regions, wherein the at least two of the two or more individual padded regions are each from the first digital image, and wherein the at least two of the two or more combined padded regions are non-overlapping;

automatically evaluate each of the two or more combined padded regions;

ignore at least one combined padded region of the two or more combined padded regions based on the evaluation of the two or more combined padded regions; and

alter the first digital image based on the combined padded region to generate a second digital image based on characteristics of an unignored combined padded region, wherein the second digital image is a recomposed version of the first digital image.

2. The computing system of claim 1 , wherein each border of the combined padded region is at least partially collinear with a border of one of the two or more individual padded regions.

3. The computing system of claim 1 , wherein the processing system is further configured to automatically alter the first digital image such that the combined padded region satisfies a broad rule of thirds.

4. The computing system of claim 1 , wherein the processing system is further configured to automatically alter the first digital image such that the combined padded region satisfies a strict rule of thirds.

5. The computing system of claim 1 , wherein the processing system is further configured to automatically alter an aspect ratio of the first digital image such that the combined padded region is preserved without removing any pixels in the combined padded region.

6. The computing system of claim 1 , wherein the processing system is further configured to automatically alter an aspect ratio of the first digital image based on a location of the combined padded region.

7. The computing system of claim 1 , wherein the processing system is further configured to determine a size of each of the two or more combined padded regions and for excluding at least one of the two or more combined padded regions from the second digital image in response to determining that the size of the at least one of the two or more combined padded regions is smaller than a largest of the two or more combined padded regions by at least a preselected magnitude.

8. The computing system of claim 7 , wherein the size of each of the two or more combined padded regions is determined by either an area or a width of each of the two or more combined padded regions.

9. The computing system of claim 1 , wherein the processing system is further configured to store, in association with the first digital image, data for identifying the combined padded region.

10. The computing system of claim 1 , wherein the processing system is further configured to define the combined padded region by combining only overlapping individual padded regions of the two or more individual padded regions.

11. The computing system of claim 10 , wherein an overlap amount of a set of overlapping individual padded regions must be greater than a preselected threshold in order to be defined as a combined padded region.

12. The computing system of claim 1 , wherein the processing system is further configured to include multiple highest scoring combined padded regions satisfying a broad rule of thirds.

13. The computing system of claim 12 , wherein the processing system is further configured to use a centroid of the multiple highest scoring combined padded regions to satisfy the broad rule of thirds.

14. The computing system of claim 12 , wherein the processing system is further configured to use a weighted centroid of the multiple highest scoring combined padded regions to satisfy the broad rule of thirds, and wherein the weighted centroid is defined by a fitness scores of the multiple highest scoring combined padded regions.

15. The computing system of claim 1 , wherein the processing system is further configured to:

receive a request for an output aspect ratio for the second digital image; and

ignore one or more lowest scoring combined padded regions at least in part in response to determining that all of the two or more combined padded regions cannot fit the requested output aspect ratio.

16. The computing system of claim 15 , wherein the processing system is further configured to automatically detect and issue a notification that a portion of the combined padded region would be deleted if the requested output aspect ratio for the second digital image was executed.

17. The computing system of claim 15 , wherein the processing system is further configured to add non-content borders to the first digital image to satisfy the requested output aspect ratio for the second digital image.

18. The computing system of claim 1 , wherein the processing system is configured to combine only overlapping individual padded regions of the two or more individual padded regions.

19. The computing system of claim 18 , wherein an overlap amount of a set of overlapping individual padded regions must be greater than a preselected threshold in order to be defined as a combined padded region.

20. The computing system of claim 1 , wherein all of two or more combined padded regions are non-overlapping.

21. The computing system of claim 1 , wherein the two or more combined padded regions comprise low-priority combined padded regions and medium-priority combined padded regions, and wherein none of the medium-priority combined padded regions overlap any other medium-priority combined padded regions.

22. The computing system of claim 1 , wherein the processing system is further configured to:

assign a fitness score to each of the two or more combined padded regions; and

ignore at least one combined padded region of the two or more combined padded regions based on the assigned fitness score of the at least one combined padded region being below a preselected threshold.

23. The computing system of claim 1 , wherein the processing system is further configured to expand the first digital image based on the combined padded region to generate the second digital image.

24. A method comprising:

storing, by a processing system, a first digital image;

identifying, by a processing system, two or more individual regions in the first digital image that each include a face;

padding, by the processing system, each of the two or more individual regions to produce two or more individual padded regions;

combining, by the processing system, at least two of the two or more individual padded regions to form two or more combined padded regions, wherein the at least two of the two or more individual padded regions are each from the first digital image, and wherein the at least two of the two or more combined padded regions are non-overlapping;

automatically evaluating, by the processing system, each of the two or more combined padded regions;

ignoring, by the processing system, at least one combined padded region of the two or more combined padded regions based on the evaluation of the two or more combined padded regions; and

altering, by the processing system, the first digital image based on the combined padded region to generate a second digital image based on characteristics of an unignored combined padded region, wherein the second digital image is a recomposed version of the first digital image.

25. The method of claim 24 , wherein all of the two or more combined padded regions are non-overlapping.

26. The method of claim 24 , further comprising:

assigning a fitness score to each of the two or more combined padded regions; and

ignoring at least one combined padded region of the two or more combined padded regions based on the assigned fitness score of the at least one combined padded region being below a preselected threshold.

27. A non-transitory computer-readable medium having instructions stored thereon that, upon execution by a computing device, cause the computing device to perform operations comprising:

storing a first digital image;

identifying two or individual regions in the first digital image that each include a face;

padding each of the two or more individual regions to produce two or more individual padded regions;

combining at least two of the two or more individual padded regions to form two or more, combined padded regions, wherein the at least two of the two or more individual padded regions are each from the first digital image, and wherein the at least two of the two or more combined added regions are non-overlapping;

automatically evaluating each of the two or more combined padded regions;

ignoring at least one combined padded region of the two or more combined padded regions based on the evaluation of the two or more combined padded regions; and

altering the first digital image based on the combined padded region to generate a second digital image based on characteristics of an unignored combined padded region, wherein the second digital image is a recomposed version of the first digital image.

28. The computing system of claim 21 , wherein the processing system is further configured to determine the fitness score for each of the two or more combined padded regions based on size, blink, gaze direction, expression, pose, occlusion, sharpness, exposure or contrast.

29. The computing system of claim 28 , wherein the processing system is further configured to automatically evaluate each of the two or more individual regions including assigning a fitness score to each of the individual regions based on size, blink, gaze direction, expression, pose, occlusion, sharpness, exposure or contrast, and for determining the fitness score for each of the two or more combined padded regions based on a weighted combination of the at least one of the two or more individual regions fitness scores.