Ranking and selecting images for display from a set of images

Patent metadata

Field	Value
Publication number	US-9858295-B2
Application number	US-201514747079-A
Country	US
Kind code	B2
Filing date	Jun 23, 2015
Priority date	Jun 24, 2014
Publication date	Jan 2, 2018
Grant date	Jan 2, 2018

Abstract

Implementations generally relate to ranking and selecting images for display from a set of images. In some implementations, a computer-implemented method includes providing selected images for display, including examining characteristics of a plurality of images, where the examined characteristics include two or more of: visual capture characteristics, visual content characteristics, and social popularity characteristics of the images. The method determines individual scores for the respective examined characteristics of the images, determines overall scores of the images based on a combination of the individual scores for the examined characteristics of the images, and determines a ranking of the images based on the overall scores. The method selects one or more images based on the ranking of the images, and causes a display of the one or more selected images.

First claim

What is claimed is:

1. A computer-implemented method to provide selected images for display, the method comprising:
examining characteristics of a plurality of images, wherein examining the characteristics includes examining visual content characteristics and one or more of: visual capture characteristics and social popularity characteristics of the plurality of images, wherein the visual content characteristics include one or more detected types of content features depicted in the plurality of images, wherein examining the one or more visual content characteristics includes:
comparing the one or more detected types of content features to a list of one or more particular content types, wherein the list includes a content type that includes human faces; and
designating for exclusion one or more particular images of the plurality of images depicting at least one type of content feature that matches one or more of the particular content types in the list;

determining individual scores for respective examined characteristics of multiple images of the plurality of images, wherein the multiple images are assigned one or more of the individual scores based on the one or more detected types of content features depicted in the multiple images;
determining overall scores of the multiple images based on a combination of the individual scores for the examined characteristics of the multiple images;
determining a ranking of the multiple images based on the overall scores;
selecting one or more images of the multiple images based on the ranking of the multiple images, wherein the one or more particular images are excluded from the selecting; and
causing a display of the one or more selected images.

2. The method of claim 1 further comprising, prior to the examining characteristics:
selecting the plurality of images as a subset of candidate images from a set of images based on an initial set of characteristics of the set of images.

3. The method of claim 2 wherein the initial set of characteristics of the set of images is operative to be processed using less computational resources than the examined characteristics of the subset of candidate images.

4. The method of claim 2 wherein the display is a public display of the one or more selected images over one or more networks, wherein the initial set of characteristics include a privacy characteristic indicating whether an associated image is allowed by an associated user to be displayed publicly, wherein selecting the plurality of images as a subset includes excluding one or more images having a privacy characteristic indicative of a restriction on public display of the one or more images.

5. The method of claim 2 wherein the initial set of characteristics includes at least one of:
a resolution of an associated image; and
noise detected in an associated image.

6. The method of claim 2 wherein selecting the plurality of images as the subset of candidate images includes detecting one or more faces depicted in the set of images, wherein images in the set of images depicting the one or more faces are excluded from the subset of candidate images.

7. The method of claim 2 wherein the set of images are provided by a plurality of different users.

8. The method of claim 1 wherein one or more of the examined characteristics each include a plurality of types of the examined characteristic, wherein determining individual scores for the respective examined characteristics of the multiple images includes:
weighting the plurality of types of the examined characteristic by different amounts; and
combining the weighted types to determine the individual score of the examined characteristic.

9. The method of claim 1 wherein a plurality of the individual scores are each weighted by a different amount in the determining of the overall scores.

10. The method of claim 1 wherein examining the characteristics of the plurality of images includes examining social popularity characteristics, including obtaining the social popularity characteristics for the plurality of images from one or more devices over a network, wherein the social popularity characteristics include one or more of:
a number of views of each of the plurality of images by one or more users of a network service,
a number of user comments for each of the plurality of images,
a share count for each of the plurality of images, and
a rating count for each of the plurality of images.

11. The method of claim 10 wherein one or more users contributing input to the network service to create the social popularity characteristics are checked for a respective associated reputation score indicating a weight of input from an associated user of the one or more users, wherein the reputation score is based on past history of input activity to the network service from the associated user, and wherein the determining of the individual scores includes weighting the social popularity characteristics based on the reputation score.

12. The method of claim 1 wherein examining the characteristics of the plurality of images includes examining social popularity characteristics, including obtaining the social popularity characteristics for the plurality of images from one or more devices over a network, wherein the social popularity characteristics include a number of times that the plurality of images have been shared between users.

13. The method of claim 12 wherein selecting one or more images based on the ranking of the multiple images includes selecting the one or more images based on one or more stored preferences for one or more of the examined characteristics, wherein the stored preferences are associated with a user associated with a device used for the display of the one or more selected images.

14. The method of claim 1 wherein the one or more types of content include text, faces, landmarks, landscape features, and objects.

15. The method of claim 1 wherein examining the characteristics includes examining the visual capture characteristics, the visual capture characteristics including at least one of: blurriness of the plurality of images and exposure level of the plurality of images.

16. A system to provide selected images for display, the system comprising:
a storage device; and
at least one processor operative to access the storage device and configured to:
determine characteristics for an image, wherein the one or more characteristics include two or more of: visual capture characteristics, visual content characteristics, and social popularity characteristics of the image;
generate one or more individual scores for the respective characteristics of the image, wherein generating one or more individual scores includes assigning a weight to each of the one or more characteristics and determining the one or more individual scores based on the one or more weighted characteristics;
combine the one or more individual scores to determine an overall score of the image, wherein combining the one or more individual scores to determine an overall score includes assigning a weight to each of the one or more individual scores and determining the overall score of the image based on the one or more weighted individual scores;
determine that the overall score satisfies a predetermined display requirement; and
cause a display of the image.

17. The system of claim 16 wherein the at least one processor is further configured to:
determine a ranking of the image with respect to one or more other images based on the overall score of the image.

18. The system of claim 16 wherei

Assignees

• Google Inc
• Google Llc

Inventors

• Murphy-Chutorian Erik
• Steiner Matthew
• Cui Jingyu

Classifications

• G06F18/24

  Classification techniques · CPC title

• G06F16/5838 Primary

  using colour · CPC title

• G06Q10/00 Primary

  Administration; Management · CPC title

• G06K9/46

  Physics · mapped topic

• G06K9/52

  Physics · mapped topic