Method for assembling a collection of digital images

• US 7,711,211 B2
• Filed: 06/08/2005
• Issued: 05/04/2010
• Est. Priority Date: 06/08/2005
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Dance, Christopher R., Snowdon, David
• Primary Examiner(s): Johns; Andrew W
• Assistant Examiner(s): TORRES, JOSE
• Application Number:
  US11/147,907
• Publication Number:
  US 20060280427A1
• Time in Patent Office: 1,791 Days
• Field of Search: 382/305
• US Class Current: 382/305
• CPC Class Codes: H04N 1/00132 in a digital photofinishing system, i.e. a system where digital photographic images undergo typical photofinishing processing, e.g. printing ordering
• CPC Class Codes: H04N 1/00143 Ordering
• CPC Class Codes: H04N 1/00161 Viewing or previewing
• CPC Class Codes: H04N 1/00167 Processing or editing H04N1/00196 - H04N1/00201 take precedence
• CPC Class Codes: H04N 1/00172 directly from a still digital camera or from a storage medium mounted in a still digital camera H04N1/0018 takes precedence
• CPC Class Codes: H04N 5/765 Interface circuits between an apparatus for recording and another apparatus
• CPC Class Codes: H04N 5/772 the recording apparatus and the television camera being placed in the same enclosure
• CPC Class Codes: H04N 5/775 between a recording apparatus and a television receiver

Abstract

A method for recommending a collection of digital images from a set of images includes specifying at least one image selection criterion. For each of a plurality of images in the set of images, an image quality value for the image is determined. Images are recommended for the collection by taking into consideration the image quality value for the images and the degree to which the collection satisfies the at least one image selection criterion.

Claims

1. A method comprising:

acquiring images from a plurality of digital image acquisition devices to form a set of images;

storing the set of images in memory of an image selection system;

specifying image selection criteria, each of the selection criteria including a weighting which reflects a relative significance of the selection criterion;

for each of a plurality of images in the set of images, determining an image quality value for the image; and

with software and hardware of the image selection system;

recommending images for a collection of digital images from the set of images including taking into consideration the image quality value for the images and the degree to which the collection satisfies the image selection criteria.

2. The method of claim 1, further comprising:

for each of a plurality of the images in the set;

extracting at least one feature within the image using a recognition component; and

labeling the image with a label which corresponds to the extracted feature, the recommending of the images including using the image labels of a plurality of the images in applying the image selection criterion.

3. The method of claim 2, wherein at least one of the selection criteria seeks to include a selected feature in the collection, the recommending of the images including selecting at least one image labeled with the feature for inclusion in the collection.

4. The method of claim 2, wherein the recommending of the images includes maximizing the number of different labeled features in the image collection.

5. The method of claim 2, wherein the features include at least one of people and locations and wherein the recognition component is operable to recognize at least one of:

a part of a person by which the person is distinguishable from other people; and

a part of a location by which the location is distinguishable from other locations.

6. The method of claim 1, wherein the method further includes interacting with a user to generate at least one of:

a number of images to be in the collection;

whether to apply automatic enhancement techniques to images in the set of images;

a feature selection criterion; and

a weighting to be applied to a feature selection criterion.

7. The method of claim 1, wherein the image selection criteria include at least one of:

attempting to include images of at least one selected person in the collection;

attempting to include images of at least one selected location in the collection;

attempting to maximizing a number of different people in the collection;

attempting to maximize a number of different locations in the collection; and

combinations thereof.

8. The method of claim 1, further including:

accessing metadata associated with images in the set; and

using the metadata to determine whether at least one of the at least one image selection criterion is satisfied.

9. The method of claim 1, wherein the recommending of images for the collection includes selecting images for the collection based on a product of the image quality value of an image and an extent to which the image satisfies the selection criteria.

10. The method of claim 1, wherein the determining of the image quality value for the image comprises automatically assigning a value to the image based on at least one of the image'"'"'s exposure, contrast, saturation, color balance, and extent of blur.

11. The method of claim 1, further comprising:

recommending a second collection of digital images from the set of images including applying at least one image selection criterion different from the first image selection criterion in recommending the second collection of images.

12. The method of claim 1, wherein the steps of determining an image quality value for the image and recommending images for the collection are automated.

13. The method of claim 1 wherein the images are selected from the group consisting of digital images, sequences comprising a plurality of images, and combinations thereof.

14. The method of claim 1, wherein the collection includes no greater than a predetermined maximum number of images which is fewer than the number of images in the set.

15. The method of claim 1, further comprising:

providing restricted access to the collection.

19. A method for forming a collection of images of an event comprising:

capturing images of the event with a plurality of digital devices;

acquiring the images captured by the digital devices with an image selection system comprising an acquisition processor; and

recommending a collection of digital images from a set of the acquired images by the method of claim 1.

16. A method comprising:

acquiring images from a plurality of digital image acquisition devices to form a set of images;

storing the set of images in memory of an image selection system;

specifying image selection criteria, each of the selection criteria including a weighting which reflects a relative significance of the selection criterion;

for each of a plurality of images in the set of images, determining an image quality value for the image; and

with software and hardware of the image selection system;

recommending images for a collection of digital images from the set of images including taking into consideration the image quality value for the images and the degree to which the collection satisfies the image selection criteria, the recommending of the images for the collection including;

a) while the collection has fewer than a desired number of images in the collection, for each image p from the set, computing a score (p) which corresponds to the change in the expected reward for the collection E[reward|C], if image p were to be included with the collection, where;

$E<br><br>\left[\mathrm{reward}|C\right]=\underset{q\in <br><br>Q}{\overset{}{\sum <br><br>}}<br><br>{\mathrm{\pi <br><br>}}_q<br><br>\underset{p\in <br><br>C}{\max }<br><br>w_q^p$

where π

_q is a probability weighting which reflects a significance of one of the selection criteria expressed by a query q of a set of queries Q, and

W_q^p, represents a measured image quality of image p that satisfies the query if the image satisfies the query q, otherwise, if the image does not satisfy query q then W_q^p=0,

C is the collection; and

b) selecting an image for the collection from a set of remaining unselected images and which has the highest score (p).

17. An article of manufacture comprising a storage medium which stores software for performing a method which includes:

providing for specifying image selection criteria;

for each of a plurality of images in a set of images, determining an image quality value for the image; and

recommending images for a collection of digital images from the set of images including taking into consideration the image quality value for the images and the degree to which the collection as a whole satisfies the image selection criteria.

18. The article of claim 17, wherein each of the provided selection criteria includes a weighting which reflects the relative significance of the selection criterion.

20. A system for recommending a collection of digital images from a set of images comprising:

a component for evaluating images in the set of images and assigning an image quality value to each of the evaluated images;

a memory which stores image selection criteria selected from;

a) a desired number or range of photographs in the collection;

b) a minimum image quality threshold for images in the collection;

c) a diversity selection criterion which specifies that the collection should attempt to capture at least one of a maximum number of different people and a maximum number of different locations;

d) a feature selection criterion, which specifies that the collection should attempt to capture at least one of the group consisting of selected people and selected locations;

e) a device or user selection criterion, which specifies that the collection should attempt to include a roughly equal number of images from each contributing device or device user; and

a module for recommending images for the collection, taking into consideration the image quality value for the images and the degree to which the collection satisfies selected ones of the image selection criteria, the module including software and hardware for selection of the images.

21. The system of claim 20, further comprising a user interface for receiving user input regarding at least one of:

a number of images to be included in the collection;

whether to apply automatic enhancement techniques to images in the set of images;

at least one selection criterion; and

a weighting of a selection criterion.

22. The system of claim 20, further comprising:

an acquisition processor for acquiring images captured by at least one digital device.

23. An assembly comprising the system of claim 20 and a plurality of digital devices, the assembly including an input device for acquiring images captured by the digital devices.

24. A system for automatically recommending a collection of digital images from a set of images comprising:

an acquisition processor for acquiring images from a plurality of digital devices;

a software component for identifying images in the set of images which are responsive to at least one of a plurality of queries and assigning a label to an image corresponding to which of the queries it is responsive to;

a software component for automatically evaluating images in the set of images for image quality;

a module including software for recommending images for the collection based on an image quality value for each image which takes into account an image'"'"'s label and its evaluated image quality; and

hardware for running the software.

25. A method comprising:

specifying image selection criteria, each of the selection criteria including a weighting which reflects a relative significance of the selection criterion;

for each of a plurality of images in the set of images, determining an image quality value for the image; and

with software and hardware of an image selection system, recommending images for a collection of digital images from the set of images including taking into consideration the image quality value for the images and the degree to which the collection satisfies the image selection criteria, the recommending of the images for the collection including;

a) white the collection has fewer than a desired number of images in the collection, for each image p remaining in the set of images which has not yet been added to the collection, computing a score (p) which corresponds to the change in an expected reward based on satisfying of the image selection criteria, assuming image p were to be included with the collection;

b) selecting an image for the collection from a set of remaining unselected images and which has the highest computed score (p); and

repeating a) and b) until the collection has the desired number of images.