System and method for intelligently determining image capture times for image applications
A method for organizing images from multiple image capture devices includes automatically determining a coarse offset between image capture times recorded in a first image capture device and image capture times recorded in a second image capture device. The coarse offset is determined by a computing a correlation between image counts of images captured by the first image capture device and images captured by the first image capture device. The method also includes adjusting the image capture times of images recorded in the second image capture device by the coarse offset to produce adjusted image capture times for images captured by the second image capture device.
1. A method for intelligently determining capture times of images for image applications, comprising:
sequencing a first group of images with known image capture times in a chronological order based on image capture times;
dividing the first group of images in the chronological order into first subclusters based on similarities between images, wherein the images in at least one of the first subclusters have a first similarity value higher than a first threshold, wherein the first similarity value is based on one or more of dominant colors, color distributions, image capture locations, detected faces, recognized faces, the number of people, sizes and/or prominence of the faces, detected objects, recognized objects, or sizes and/or prominence of objects in the images, wherein each of the first subclusters comprises images from the first group having adjacent image capture times;
dividing a second group of images with unknown image capture times into second subclusters based on similarities between images;
computing, by a computer system, a second similarity value between a first subcluster of images with known image capture times and a second subcluster of images with unknown image capture times; and
assigning an image capture time of the first subcluster of images to the second subcluster of images if the second similarity value between the first subcluster of images and the second subcluster of images is above a second threshold, thereby producing a unified image capture times for the first subcluster of images to the second subcluster of images based on known image capture times of first subclusters and image capture times assigned to the second subclusters.
2. The method of claim 1 , further comprising:
sequencing the second group of images before the step of dividing a second group of images into second subclusters, wherein the second subclusters each comprise images adjacent in the sequence.
3. The method of claim 2 , wherein the images in the second group are sequenced based on a parameter that is not image capture time.
4. The method of claim 3 , wherein the images in the second group are sequenced based on a parameter selected from the group consisting of file names, image upload times, image reception times, and image processing times.
5. The method of claim 1 , wherein the images in one of the second subclusters have similarities higher than a third threshold.
6. The method of claim 1 , wherein the step of calculating a second similarity value between a first subcluster of images and the second subcluster of images comprises:
calculating differences in one or more of parameters between the images in the first subcluster and the images the second subcluster, wherein the second similarity value is inversely related to the differences, wherein the one or more of parameters comprise one or more of dominant colors, color distributions, image capture locations, detected faces, recognized faces, the number of people, sizes and/or prominence of the faces, detected objects, recognized objects, or sizes and/or prominence of objects in the images.
7. The method of claim 1 , wherein the step of computing, by a computer system, a second similarity value between a first subcluster of images with known image capture times and a second subcluster of images with unknown image capture times comprises:
calculating differences of a set of parameters between the first subcluster of images and the second subcluster of images;
setting up a topological space based on the set of parameters; and
calculating a topological distance in the topological space between the first subcluster of images and the second subcluster of images based on the calculated differences, wherein the second similarity value is inversely related to the topological distance.
8. The method of claim 1 , wherein the first group of images with known image capture times are obtained by multiple image capture devices, the method further comprising:
adjusting, by an offset, image capture times recorded by at least one of the multiple image capture devices.
9. The method of claim 8 , wherein the step of assigning comprising:
assigning a unified image capture time of the first subcluster of images to the second subcluster of images if the second similarity value between the first subcluster of images and the second subcluster of images is above the second threshold, the method further comprising:
characterizing the first group of images by unified image capture times which adjust offset(s) between image capture times recorded by different ones of the multiple image capture devices.
10. The method of claim 1 , further comprising:
sequencing the first group of images and the second group of images in a chronological order based on the unified image capture times.
11. The method of claim 10 , further comprising:
incorporating at least some of the first group of images and the second group of images in the chronological order into a design of an image product.
12. The method of claim 11 , wherein the image product comprises a photobook, a photo calendar, a photo story, a photo blog, or a photo slideshow.
13. The method of claim 10 , further comprising:
displaying, sharing or publishing at least some of the first group of images and the second group of images in the chronological order.
14. The method of claim 1 , further comprising:
incorporating at least some of the second group of images into a design of an image product based on the image capture times assigned to the second group of images.
15. The method of claim 14 , wherein the image product comprises a photobook, a photo calendar, a photo story, a photo blog, or a photo slideshow.
16. The method of claim 1 , further comprising:
displaying, sharing or publishing at least some of the second group of images based on the image capture times assigned to the second group of images.
17. The method of claim 1 , further comprising:
tagging the second group of images by the image capture times assigned to the second group of images.
18. The method of claim 17 , further comprising:
searching or categorizing the second group of images based on the image capture times assigned to the second group of images.
19. A method for organizing images from multiple image capture devices, comprising:
computing, by a computer system, a correlation function over image capture time between image counts of images captured by a first image capture device and image counts of images captured by a second image capture device, wherein the step of computing further comprising:
sequencing images captured by the first image capture device based on the image capture times recorded by the first image capture device;
sampling image counts of images captured by the first image capture device to produce a first image count distribution (ICD);
sequencing images captured by the second image capture device based on the image capture times recorded by the second image capture device;
sampling image counts of images captured by the second image capture device to produce a second ICD;
determining using the correlation function, by the computer system, a coarse offset between image capture times recorded in the first image capture device and image capture times recorded in the second image capture device, wherein the step of computing further determining:
computing a correlation function between the first ICD and the second ICD by the computer system;
identifying, in the correlation function, one or more correlation peaks that are above a correlation threshold; and
obtaining one or more coarse offsets from the one or more correlation peaks; and
subtracting the image capture times of images recorded in the second image capture device by the one of the one or more coarse offsets to produce adjusted image capture times for images captured by the second image capture device.
20. The method of claim 19 , further comprising:
for the images associated with one of the one or more correlation peaks above the correlation threshold, computing a similarity value between images captured by the first image capture device and by the second image capture device by the computer system;
automatically determining a fine offset between the image capture times recorded in the first image capture device and the image capture times recorded in the second image capture device if the similarity value is above a similarity threshold value; and
adjusting the image capture times of the images captured by the second image capture device by the fine offset.
21. The method of claim 20 , wherein the step of calculating a similarity value between images captured by the first image capture device and by the second image capture device comprises:
calculating differences in one or more of parameters between the images captured by the first image capture device and by the second image capture device, wherein the similarity value is inversely related to the differences, wherein the one or more of parameters comprise colors, color distributions, textures, image capture locations, faces and bodies detected in the images, recognized faces, the number of people, texture and colors of hair and hats if found in the images, sizes and/or prominence of the faces, detected objects, recognized objects, or sizes and/or prominence of objects in the images.