System and method for collaborative image selection

Abstract

A system and method for sharing images and collaborating in the selection of images likely to be interest to a user. Images (e.g., photos) are shared within a community of users, by allowing user to access any unrestricted community image, recommend an image to another user, assemble compilations of any images the user can access, etc. Various metrics regarding user activity are tracked, such as how often an image was viewed, downloaded, recommended, included in a compilation, printed, edited, etc. The metric values may be normalized, and then weighted and combined to produce, for each image, a ranking or rating personalized to an individual user. Ratings of images for a user may be used to order them for electronic presentation, assemble a set of images for an electronic compilation (e.g., an online album) or physical compilation (e.g., a photo book, a yearbook) or for some other purpose.

1 . A method of selecting images for a personalized compilation of images, the method comprising: 
 maintaining a pool of images for a community of users;    weighting each of the images based on characteristics of said image, including: 
 image quality; and  
 image clarity;  
   further weighting each of the images for each user based on said user's interaction with said image;    inferring a relationship between multiple users in the community of users based on said users' interactions with a subset of images in the pool of images;    adjusting said image weightings for each of the multiple users based on the related users' interaction with the subset of images; and    selecting images for a personalized compilation for a given user of the community of users based on said weightings.    
     
     
         2 . The method of  claim 1 , wherein a user's interaction with an image includes one or more of: 
 a duration of time spent viewing the image;    a frequency with which the image is viewed; and    a recommendation of the image to another user.    
     
     
         3 . The method of  claim 1 , further comprising: 
 further weighting an image for a user based on the exclusivity of said user's interaction with said image, wherein the user's interaction with an image is exclusive if no other user in the community of users has the same interaction with said image.    
     
     
         4 . The method of  claim 1 , further comprising: 
 further weighting an image for a user based on the exclusivity of said user's interaction with said image, wherein said exclusivity comprises viewing a version of the image not viewed by other users in the community.    
     
     
         5 . The method of  claim 1 , further comprising: 
 further weighting an image for a user based on the exclusivity of said user's interaction with said image, wherein said exclusivity comprises receiving one of a limited number of recommendations of the image from another user within the community.    
     
     
         6 . The method of  claim 1 , wherein said inferring comprises: 
 determining that the multiple users had identical interaction with the subset of images.    
     
     
         7 . The method of  claim 1 , wherein said inferring comprises: 
 determining that the multiple users had similar interaction with the subset of images.    
     
     
         8 . The method of  claim 7 , wherein said similar interaction comprises one or more of: 
 viewing the subset of images;    printing the subset of images;    receiving a recommendation regarding the subset of images; and    corresponding regarding the subset of images.    
     
     
         9 . The method of  claim 1 , further comprising: 
 automatically determining a subject of a first image in the pool; and    weighting the first image based on weights associated with other images containing the subject.    
     
     
         10 . A computer readable medium storing instructions that, when executed by a computer, cause the computer to perform a method of selecting images for a personalized compilation of images, the method comprising: 
 maintaining a pool of images for a community of users;    weighting each of the images based on characteristics of said image, including: 
 image quality; and  
 image clarity;  
   further weighting each of the images for each user based on said user's interaction with said image;    inferring a relationship between multiple users in the community of users based on said users' interactions with a subset of images in the pool of images;    adjusting said image weightings for each of the multiple users based on the related users' interaction with the subset of images; and    selecting images for a personalized compilation for a given user of the community of users based on said weightings.    
     
     
         11 . A computerized system for compiling personalized sets of images for different users, comprising: 
 storage means for maintaining a pool of images available for access by a community of users;    recording means for recording activity regarding the images, including: 
 a duration of time a user views an image;  
 a frequency with which a user views an image;  
 addition of a caption to an image;  
 requesting an enlarged view of a viewed image; and  
 recommendation of an image to a single user by multiple other users;  
   means for determining possible subjects of one or more images by noting: 
 which users access said image;  
 which users, if any, said image is recommended to by the users who access said image; and  
 any relationships between the users who access said image and users to whom they recommend said image; and  
   means for automatically compiling personalized sets of images likely to be of interest to the different users.