System and method for reducing similar photos for display and product design

1. A computer-implemented method for reducing similar photos for display and product design, comprising:

calculating hash values of images, wherein the images includes a first image and a second image;

calculating similarity distances between the images using at least the hash values by a similarity distance computation module in the computer system;

automatically grouping the images into a burst by a burst grouping module in the computer system when the similarity distances of the images are below a burst threshold value;

designating the first image as a duplicate image if a similarity distance between the first image and the second image is below a duplicate threshold value, wherein the duplicate threshold value is smaller than the burst threshold value;

designating the first image as a new image if similarity distances between the first image and other images in the burst are above the duplicate threshold value, wherein the new image is assigned to the burst when the similarity distances between the new image and other images in the burst are below the burst threshold value;

automatically selecting at least one image from the burst of images; and

automatically creating a photo product design using the selected image from the burst by an intelligent design creation engine in the computer system.

2. The computer-implemented method of claim 1 , wherein the duplicate image is discarded before or in step of selecting at least one image from the burst of images.

3. The computer-implemented method of claim 1 , further comprising:

sequencing the images in a chronological sequence, wherein the similarity distances are calculated between adjacent images in the chronological sequence.

4. The computer-implemented method of claim 1 , wherein the images are grouped into a burst based further on at least one of image capture times or image capture locations.

5. The computer-implemented method of claim 1 , wherein the images are automatically grouped into the burst further based on image compositions, facial expressions, orientations of the faces, sizes of the faces, light exposures of the faces and persons in the image, importance of person(s) in the image, scene composition, dominant colors, color histograms, block color histograms, or cloth colors and patterns.

6. The computer-implemented method of claim 1 , wherein at least one image is automatically selected from the burst of images based on detection of faces, faces recognized presence of faces of family and close friends, or positions and focus of the faces in the image.

7. The computer-implemented method of claim 1 , further comprising:

automatically selecting a product type, a product layout, or a product style for the photo product design by the intelligent design creation engine.

8. The computer-implemented method of claim 1 , further comprising:

automatically presenting multiple images in a burst by the intelligent design creation engine for user selection of a different image from the burst in the photo product design or the photo display.

9. The computer-implemented method of claim 1 , further comprising:

automatically displaying the selected image from the burst on a device by the intelligent design creation engine.

10. The computer-implemented method of claim 1 , wherein step of calculating similarity distances comprises:

automatically calculating a Hamming Distance between hash values of the first image and the second image.

11. The computer-implemented method of claim 10 , wherein the hash values of the first image and the second image are calculated by a hashing method including pHash, Radial-Hash, or Wavelet-based hash.

12. A computer-implemented method for reducing similar photos for display and product design, comprising:

calculating hash values of images by a similarity distance computation module in a computer system, wherein the images includes a first image and a second image;

calculating similarity distances between the images using at least the hash values by a similarity distance computation module in the computer system, wherein step of calculating similarity distances further includes:

calculating differences in color densities, edge orientations, or edge densities between the images;

automatically grouping the images into a burst based at least in part on the similarity distances of the images by a burst grouping module in the computer system;

automatically selecting at least one image from the burst of images, wherein automatically selecting at least one image from the burst of images comprises:

designating the first image as a duplicate image if a similarity distance between the first image and the second image is below a duplicate threshold value, wherein the duplicate image is discarded before step of selecting and the duplicate threshold value is smaller than the burst threshold value; and

automatically creating a photo product design using the selected image from the burst by an intelligent design creation engine in the computer system.

13. The computer-implemented method of claim 12 , wherein the images are grouped into the burst when the similarity distances of the images are below a burst threshold value.

14. A computer-implemented method for reducing similar photos for display and product design, comprising:

calculating hash values of images by a similarity distance computation module in a computer system, wherein the images includes a first image and a second image;

calculating similarity distances between the images using at least the hash values by a similarity distance computation module in the computer system;

automatically grouping the images into a burst based at least in part on the similarity distances of the images by a burst grouping module in the computer system, wherein automatically grouping the images comprises:

designating the first image as a new image if similarity distances between the first image and other images in the images are above the duplicate threshold value, wherein the new image is assigned into the burst when the similarity distances between the new image and other images in the burst are below the burst threshold value;

automatically selecting at least one image from the burst of images based in part on detection of faces, faces recognized presence of faces of family and close friends, or positions and focus of the faces in the image; and

automatically creating a photo product design using the selected image from the burst by an intelligent design creation engine in the computer system.

15. The computer-implemented method of claim 14 , wherein step of automatically selecting comprises:

designating the first image as a duplicate image if a similarity distance between the first image and the second image is below a duplicate threshold value, wherein the duplicate image is discarded before step of selecting.

16. A computer-implemented method for reducing similar photos for display and product design, comprising:

calculating hash values of images, wherein the images includes a first image and a second image;

calculating similarity distances between the images using at least the hash values by a similarity distance computation module in the computer system;

automatically grouping the images into a burst by a burst grouping module in the computer system when the similarity distances of the images are below a burst threshold value, wherein the images are automatically grouped into the burst further based on image compositions, facial expressions, orientations of the faces, sizes of the faces, light exposures of the faces and persons in the image, importance of person(s) in the image, scene composition, dominant colors, color histograms, block color histograms, or cloth colors and patterns;

designating the first image as a duplicate image if a similarity distance between the first image and the second image is below a duplicate threshold value, wherein the duplicate threshold value is smaller than the burst threshold value;

automatically selecting at least one image from the burst of images; and

automatically creating a photo product design using the selected image from the burst by an intelligent design creation engine in the computer system.

17. The computer-implemented method of claim 16 , wherein the duplicate image is discarded before or in step of selecting at least one image from the burst of images.

18. The computer-implemented method of claim 16 , further comprising:

sequencing the images in a chronological sequence, wherein the similarity distances are calculated between adjacent images in the chronological sequence.

19. The computer-implemented method of claim 16 , wherein the images are grouped into a burst based further on at least one of image capture times or image capture locations.

20. The computer-implemented method of claim 16 , wherein at least one image is automatically selected from the burst of images based on detection of faces, faces recognized presence of faces of family and close friends, or positions and focus of the faces in the image.

21. The computer-implemented method of claim 16 , further comprising:

automatically selecting a product type, a product layout, or a product style for the photo product design by the intelligent design creation engine.

22. The computer-implemented method of claim 16 , further comprising:

automatically presenting multiple images in a burst by the intelligent design creation engine for user selection of a different image from the burst in the photo product design or the photo display.

23. The computer-implemented method of claim 16 , further comprising:

automatically displaying the selected image from the burst on a device by the intelligent design creation engine.

24. The computer-implemented method of claim 16 , wherein step of calculating similarity distances comprises:

automatically calculating a Hamming Distance between hash values of the first image and the second image.

25. The computer-implemented method of claim 24 , wherein the hash values of the first image and the second image are calculated by a hashing method including pHash, Radial-Hash, or Wavelet-based hash.