Generating a reference digital image based on an indicated time frame and searching for other images using the reference digital image

Patent metadata

Field	Value
Publication number	US-9836669-B2
Application number	US-201615049918-A
Country	US
Kind code	B2
Filing date	Feb 22, 2016
Priority date	Feb 22, 2016
Publication date	Dec 5, 2017
Grant date	Dec 5, 2017

Abstract

A digital image and a text string is received. The text string can be processed to identify at least a time frame and determine whether the time frame is a future time frame or a past time frame. How at least one element of the first digital image will change or has changed during the time frame can be predicted. At least one reference digital image can be generated, the reference digital image including at least one change to the at least one element corresponding to how the at least one element will change or has changed during the time frame. The reference digital image to each of a plurality of other digital images. A correlation parameter can be assigned to each of the plurality of other digital images. A portion of the plurality of other digital images having highest correlation parameters can be output for presentation to a user.

First claim

What is claimed is:

1. A method comprising:
receiving a first digital image;
receiving a text string indicating at least a time frame;
processing the text string to identify at least the time frame and determine whether the time frame is a future time frame or a past time frame;
responsive to determining whether the time frame is the future time frame or the past time frame, predicting how at least one element of the first digital image will change or has changed during the time frame;
responsive to predicting how the at least one element of the first digital image will change or has changed during the time frame, generating, using a processor, at least one reference digital image that is a revised version of the first digital image, the reference digital image including at least one change to the at least one element corresponding to how the at least one element will change or has changed during the time frame;
comparing the reference digital image to each of a plurality of other digital images;
based on comparing the reference digital image to each of the plurality of other digital images, assigning a correlation parameter to each of the plurality of other digital images, each correlation parameter indicating how closely a respective other digital image matches the reference digital image; and
outputting for presentation to a user a portion of the plurality of other digital images having highest correlation parameters.

2. The method of claim 1 , wherein:
the text string further indicates a name of a person; and
the correlation parameter is assigned to a respective other digital image is based, at least in part, on whether the other digital image is associated with the name of the person indicated in the text string.

3. The method of claim 1 , wherein:
the text string further indicates a name of a location; and
the correlation parameter is assigned to a respective other digital image is based, at least in part, on whether the other digital image is associated with location indicated in the text string.

4. The method of claim 1 , wherein processing the text string to identify at least the time frame and determine whether the time frame is a future time frame or a past time frame comprises the text string using natural language processing.

5. The method of claim 1 , wherein outputting for presentation to the user the portion of the plurality of other digital images having highest correlation parameters comprises presenting the portion of the plurality of other digital images having highest correlation parameters in an order based the correlation parameters.

6. The method of claim 1 , further comprising:
assigning a ranking to each of the portion of the plurality of other digital images having highest correlation parameters;
wherein outputting for presentation to the user the portion of the plurality of other digital images having highest correlation parameters comprises presenting the portion of the plurality of other digital images having highest correlation parameters in an order based the assigned rankings.

7. The method of claim 1 , wherein the other digital images are accessed from a social networking system.

8. A system, comprising:
a processor programmed to initiate executable operations comprising:
receiving a first digital image;
receiving a text string indicating at least a time frame;
processing the text string to identify at least the time frame and determine whether the time frame is a future time frame or a past time frame;
responsive to determining whether the time frame is the future time frame or the past time frame, predicting how at least one element of the first digital image will change or has changed during the time frame;
responsive to predicting how the at least one element of the first digital image will change or has changed during the time frame, generating at least one reference digital image that is a revised version of the first digital image, the reference digital image including at least one change to the at least one element corresponding to how the at least one element will change or has changed during the time frame;
comparing the reference digital image to each of a plurality of other digital images;
based on comparing the reference digital image to each of the plurality of other digital images, assigning a correlation parameter to each of the plurality of other digital images, each correlation parameter indicating how closely a respective other digital image matches the reference digital image; and
outputting for presentation to a user a portion of the plurality of other digital images having highest correlation parameters.

9. The system of claim 8 , wherein:
the text string further indicates a name of a person; and
the correlation parameter is assigned to a respective other digital image is based, at least in part, on whether the other digital image is associated with the name of the person indicated in the text string.

10. The system of claim 8 , wherein:
the text string further indicates a name of a location; and
the correlation parameter is assigned to a respective other digital image is based, at least in part, on whether the other digital image is associated with location indicated in the text string.

11. The system of claim 8 , wherein processing the text string to identify at least the time frame and determine whether the time frame is a future time frame or a past time frame comprises the text string using natural language processing.

12. The system of claim 8 , wherein outputting for presentation to the user the portion of the plurality of other digital images having highest correlation parameters comprises presenting the portion of the plurality of other digital images having highest correlation parameters in an order based the correlation parameters.

13. The system of claim 8 , the executable operations further comprising:
assigning a ranking to each of the portion of the plurality of other digital images having highest correlation parameters;
wherein outputting for presentation to the user the portion of the plurality of other digital images having highest correlation parameters comprises presenting the portion of the plurality of other digital images having highest correlation parameters in an order based the assigned rankings.

14. The system of claim 8 , wherein the other digital images are accessed from a social networking system.

15. A computer program product comprising a computer readable storage medium having program code stored thereon, the program code executable by a processor to perform a method comprising:
receiving, by the processor, a first digital image;
receiving, by the processor, a text string indicating at least a time frame;
processing, by the processor, the text string to identify at least the time frame and determine whether the time frame is a future time frame or a past time frame;
responsive to determining whether the time frame is the future time frame or the past time frame, predicting, by the processor, how at least one element of the first digital image will change or has changed during the time frame;
responsive to predicting how the at least one element of the first digital image will change or has changed during the time frame, generating, by the processor, at least one reference digital image that is a revised version of the first digital image, the reference digital image including at least one change to the at least one element corresponding to how the at least one element will change or has changed during the time frame;
comparing, by the processor, the reference digital image to each of a

Assignees

• IBM

Inventors

• Abou Mahmoud Alaa
• Bastide Paul R
• Lu Fang

Classifications

• G06V20/70

  Labelling scene content, e.g. deriving syntactic or semantic representations · CPC title

• G06V10/75

  Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries · CPC title

• G06F2218/12

  Classification; Matching · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title

• G06T2207/20076

  Probabilistic image processing · CPC title