Selection of representative images

• US 9,367,756 B2
• Filed: 04/16/2014
• Issued: 06/14/2016
• Est. Priority Date: 08/31/2010
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Pillai, Anand, Rabinovich, Andrew
• Primary Examiner(s): Carter, Aaron W
• Application Number:
  US14/254,242
• Publication Number:
  US 20150169978A1
• Time in Patent Office: 790 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 18/23 Clustering techniques
• CPC Class Codes: G06F 18/232 Non-hierarchical techniques
• CPC Class Codes: G06V 10/763 Non-hierarchical techniques, e.g. based on statistics of modelling distributions
• CPC Class Codes: G06V 40/172 Classification, e.g. identification

Abstract

Methods and systems for selecting a representative image of an entity are disclosed. According to one embodiment, a computer-implemented method for selecting a representative image of an entity is disclosed. The method includes: accessing a collection of images of the entity; clustering, based on similarity of one or more similarity features, images from the collection to form a plurality of similarity clusters; and selecting the representative image from one of said similarity clusters. Further, based on cluster size of said similarity clusters popular clusters can be determined, and the selection of the representative image can be from the popular clusters. In addition, the method can further include assigning a headshot score based upon a portion of the respective image covered by the entity to respective images in said popular clusters, and further selecting the representative image based upon the headshot score.

Claims

1. A computer-implemented method comprising:

obtaining a collection of images of a particular object;

for each image of the collection, identifying a sub-area of the image that (i) is centered around the particular object, and (ii) has a predetermined size or aspect ratio;

generating a score for each of two or more of the images of the collection using representative image selection criterion that include image feature coverage criteria relating to an amount of the sub-area of the image that is occupied by a predetermined feature of the particular object; and

selecting a particular image of the collection as a representative image of the particular object based on the scores.

2. The method of claim 1, wherein obtaining a collection of images of a particular object comprises:

determining that obtained images are images of the particular object; and

in response to determining that the obtained images are images of the particular object, including the images in the collection of images.

3. The method of claim 1, wherein generating a score for each of two or more of the images of the collection using representative image selection criterion that include image feature coverage criteria relating to an amount of the sub-area of the image that is occupied by a predetermined feature of the particular object, comprises:

detecting the particular object in a sub-area of a first image; and

generating a score for the first image based on at least an amount of the sub-area of the first image that is occupied by the detected particular object.

4. The method of claim 1, wherein generating a score for each of two or more of the images of the collection using representative image selection criterion comprises:

obtaining a required aspect ratio;

determining a rectangular area in a first image including the particular object;

determining margins based on the required aspect ratio and a size of the rectangular area; and

generating a score for the first image based on at least the determined margins and the size of the determined rectangular area.

5. The method of claim 1, wherein selecting a particular image of the collection as a representative image based on the scores comprises:

clustering the images of the particular object into image clusters based on image feature similarity; and

selecting the particular image of the collection as the representative image based on the scores and quantities of images in the image clusters.

6. The method of claim 5, wherein selecting the particular image of the collection as the representative image based on the scores and quantities of images in the image clusters comprises:

determining a particular image cluster includes more images than other image clusters; and

in response to determining the particular image cluster includes more images than the other image clusters, selecting an image of the particular image cluster that includes a highest score as the representative image.

7. The method of claim 1, wherein the particular object comprises a face of a person.

16. The method of claim 1, wherein generating a score for each of two or more of the images of the collection using representative image selection criterion that include image feature coverage criteria relating to an amount of the sub-area of the image that is occupied by a predetermined feature of the particular object, comprises:

generating a first score for a first image of the collection based at least on an amount of the sub-area of the first image that is occupied by the predetermined feature of the particular object; and

generating a second score for a second image of the collection based at least on an amount of a sub-area of the second image that is occupied by the predetermined feature of the particular object.

17. The method of claim 1, comprising:

determining to provide a representative image of the particular object; and

in response to determining to provide the representative image of the particular object, providing the particular image selected as the representative image of the particular object without providing other images of the collection of images of the particular object.

8. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising;

obtaining a collection of images of a particular object;

for each image of the collection, identifying a sub-area of the image that (i) is centered around the particular object, and (ii) has a predetermined size or aspect ratio;

generating a score for each of two or more of the images of the collection using representative image selection criterion that include image feature coverage criteria relating to an amount of the sub-area of the image that is occupied by a predetermined feature of the particular object; and

selecting a particular image of the collection as a representative image of the particular object based on the scores.

9. The system of claim 8, wherein obtaining a collection of images of a particular object comprises:

determining that obtained images are images of the particular object; and

in response to determining that the obtained images are images of the particular object, including the images in the collection of images.

10. The system of claim 8, wherein generating a score for each of two or more of the images of the collection using representative image selection criterion that include image feature coverage criteria relating to an amount of the sub-area of the image that is occupied by a predetermined feature of the particular object, comprises:

detecting the particular object in a sub-area of a first image; and

generating a score for the first image based on at least an amount of the sub-area of the first image that is occupied by the detected particular object.

11. The system of claim 8, wherein generating a score for each of two or more of the images of the collection using representative image selection criterion comprises:

obtaining a required aspect ratio;

determining a rectangular area in a first image including the particular object;

determining margins based on the required aspect ratio and a size of the rectangular area; and

generating a score for the first image based on at least the determined margins and the size of the determined rectangular area.

12. The system of claim 8, wherein selecting a particular image of the collection as a representative image based on the scores comprises:

clustering the images of the particular object into image clusters based on image feature similarity; and

selecting the particular image of the collection as the representative image based on the scores and quantities of images in the image clusters.

13. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:

obtaining a collection of images of a particular object;

for each image of the collection, identifying a sub-area of the image that (i) is centered around the particular object, and (ii) has a predetermined size or aspect ratio;

generating a score for each of two or more of the images of the collection using representative image selection criterion that include image feature coverage criteria relating to an amount of the sub-area of the image that is occupied by a predetermined feature of the particular object; and

selecting a particular image of the collection as a representative image of the particular object based on the scores.

14. The medium of claim 13, wherein obtaining a collection of images of a particular object comprises:

determining that obtained images are images of the particular object; and

in response to determining that the obtained images are images of the particular object, including the images in the collection of images.

15. The medium of claim 13, wherein generating a score for each of two or more of the images of the collection using representative image selection criterion that include image feature coverage criteria relating to an amount of the sub-area of the image that is occupied by a predetermined feature of the particular object, comprises:

detecting the particular object in a sub-area of a first image; and

generating a score for the first image based on at least an amount of the sub-area of the first image that is occupied by the detected particular object.