Image recomposition from face detection and facial features

1. A system comprising:

a memory; and

a processing system coupled to the memory and configured to:

identify two or more individual regions in a digital image that each include a body part, wherein each of the two or more individual regions has a respective width and a respective height, and wherein the digital image has a width and a height;

pad a first of the two or more individual regions, wherein an amount of padding added to the first of the two or more individual regions is based on a percentage of a greater of the width or the height of the first of the two or more individual regions compared to a lesser of the width or the height of the digital image;

pad a second of the two or more individual regions, wherein an amount of padding added to the second of the two or more individual regions is based on a percentage of a greater of the width or the height of the second of the two or more individual regions compared to the lesser of the width or the height of the digital image; and

define a combined padded region that includes the two or more individual padded regions.

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

3. The system of claim 1 , wherein the processing system is further configured to modify the digital image such that the combined padded region satisfies a. rule of thirds composition.

4. The system of claim 1 , wherein the rule of thirds composition is based on an adjustable parameter that specifies an amount of expansion of an individual region.

5. The system of claim 1 , wherein the processing system is further configured to modify an aspect ratio of the digital image, and wherein to modify the aspect ratio the processor is configured to delete a portion of the digital image such that the combined padded region is preserved without removing any of its pixels.

6. The system of claim 1 , wherein the processing system is further configured to modify an aspect ratio of the digital image such that a portion of only one of the two or more individual padded regions is deleted, responsive to determining that modifying the aspect ratio of the digital image cannot be completed together with preserving the combined padded region.

7. The system of claim 1 , wherein the processing system is further configured to detect and issue a notification that a portion of the combined padded region would be deleted if a preselected modification of the digital image is executed.

8. The system of claim 1 , wherein the processing system is further configured to determine a size of each of the two or more individual padded regions, and wherein to determine the size of each of the two or more individual padded regions the processor is configured to exclude at least one of the two or more individual padded regions from the combined padded region if the size of the at least one of the two or more individual padded regions is smaller than a largest one of the two or more individual padded regions by a preselected magnitude.

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

evaluate each of the two or more individual regions to assign a fitness score to each of the two or more individual regions; and

ignore at least one of the two or more individual regions whose fitness score is below a preselected threshold, wherein the combined padded region includes two or more individual regions whose fitness score is above the preselected threshold.

10. The system of claim 9 , wherein a fitness score for a particular region is based on a facial size, a facial location, an eye blink, a gaze direction, an expression, a pose, an occlusion, sharpness, exposure or contrast of a body part corresponding to the particular region.

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

12. The system of claim 1 , wherein the processing system is further configured to zoom in or out of the combined padded region by a preselected amount to form a modified digital image.

13. The system of claim 1 , wherein the processing system is further configured to modify an aspect ratio of the digital image such that equal amounts of the individual padded region are cropped from opposing sides of the padded region, responsive to determining that modifying the aspect ratio of the digital image cannot be completed together with preserving the padded region without removing any of its pixels.

14. The system of claim 13 , wherein the processing system is further configured to detect that a portion of the individual padded region would be cropped if a preselected modification of the digital image is executed.

15. The system of claim 1 , wherein the processing system is further configured to determine a width of one of the two or more individual regions; and

determine a size of padding corresponding to the one of the two or more individual regions based on the width of the one of the two or more individual regions.

16. The system of claim 1 , wherein the processing system is further configured to modify an aspect ratio of the digital image such that about forty percent of a total amount cropped from the digital image is cropped from a top border of the digital image, responsive to determining that modifying the aspect ratio of the digital image cannot be completed together with preserving the individual padded region without removing any of its pixels.

17. The system of claim 1 , wherein the processing system is further configured to modify an aspect ratio of the digital image such that equal amounts of the combined padded region are cropped from opposing sides of the combined padded region, responsive to determining that modifying the aspect ratio of the digital image cannot be completed together with preserving the combined padded region without removing any of its pixels.

18. The system of claim 1 , wherein the processing system is further configured to modify an aspect ratio of the digital image such that the combined padded region is centered in the digital image between at least one pair of opposing borders of the digital image.

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

determine a width of at least one of the two or more individual regions; and

determine a size of padding corresponding to the at least one of the two or more individual regions based on the width of the at least one of the two or more individual regions.

20. The system of claim 1 , wherein the amount of padding added is non-linearly related to the percentage.

21. The system of claim 1 , wherein a border of the combined padded region is collinear with a border of one of the two or more individual padded regions.

22. The system of claim 1 , wherein the two or more individual regions are padded such that smaller individual regions receive a greater amount of padding than larger individual regions.

23. The system of claim 1 , wherein the amount of padding added to the first of the two or more individual regions and the amount of padding added to die second of the two or more individual regions is further based on a desired image aspect ratio for a final image product.

24. A method comprising:

identifying, by a computing device, two or more individual regions in a digital image that each include a body part, wherein each of the two or more individual regions has a respective width and a respective height, and wherein the digital image has a width and a height;

padding, by the computing device, a first of the two or more individual regions, wherein an amount of padding added to the first of the two or more individual regions is based on a percentage of a greater of the width or the height of the first of the two or more individual regions compared to a lesser of the width or the height of the digital image; and

padding, by the computing device, a second of the two or more individual regions, wherein an amount of padding added to the second of the two or more individual regions is based on a percentage of a greater of the width or the height of the second of the two or more individual regions compared to the lesser of the width or the height of the digital image;

defining, by the computing device, a combined padded region that includes the two or more individual padded regions.

25. The method of claim 24 , further comprising modifying an aspect ratio of the digital image, wherein the modifying comprises deleting a portion of the digital image such that the combined padded region is preserved without removing any of its pixels.

26. A non-transitory computer-readable medium having instructions stored thereon, the instructions comprising:

instructions to identify two or more individual regions in a digital image that each include a body part, wherein each of the two or more individual regions has a respective width and a respective height, and wherein the digital image has a width and a height;

instructions to pad a first of the two or more individual regions, wherein an amount of padding added to the first of the two or more individual regions is based on a percentage of a greater of the width or the height of the first of the two or more individual regions compared to a lesser of the width or height of the digital image; and

instructions to pad a second of the two or more individual regions, wherein an amount of padding added to the second of the two or more individual regions is based on a percentage of a greater of the width or the height of the second of the two or more individual regions compared to the lesser of the width or the height of the digital image;

instructions to define a combined padded region that includes the two or more individual padded regions.

27. The non-transitory computer-readable medium of claim 26 , the instructions further comprising instructions to modify an aspect ratio of the digital image, wherein the modifying comprises deleting a portion of the digital image such that the combined padded region is preserved without removing any of its pixels.