Systems and Methods of Building and Using an Image Catalog

Abstract

A method executes at a server system to manage an image catalog. The system receives reduced-resolution versions of images stored in an image database of an external service distinct from the system, and receives a specified subject matter for the images. For each of the received reduced-resolution versions, the system creates a respective index entry in the image catalog. The respective index entry comprises keywords extracted from the respective reduced-resolution version by performing semantic analysis on the respective reduced-resolution version using a convolutional neural network trained on images corresponding to the specified subject matter. The system receives a user query and matches the query to an index entry in the image catalog. The index entry was created for a reduced-resolution version of a full-resolution image stored in the image database. The system retrieves the full-resolution image from the image database and transmits the first full-resolution image to the user.

What is claimed is: 
     
         1 . A method of managing an image catalog, performed by one or more servers, each having one or more processors and memory, the method comprising, at the one or more servers:
 receiving reduced-resolution versions of one or more images stored in an image database of an external service distinct from the one or more servers, and receiving a specified subject matter for the one or more images;   for each of the received reduced-resolution versions, creating a respective index entry in the image catalog, the respective index entry comprising one or more keywords extracted from the respective reduced-resolution version by performing semantic analysis on the respective reduced-resolution version using a convolutional neural network trained on images corresponding to the specified subject matter;   receiving a query from a user;   matching the query to a first index entry in the image catalog, the first index entry created for a first reduced-resolution version of a first full-resolution image stored in the image database of the external service;   retrieving the first full-resolution image from the image database of the external service; and   transmitting the first full-resolution image to the user.   
     
     
         2 . The method of  claim 1 , wherein the first index entry comprises a respective color palette for the first reduced-resolution version. 
     
     
         3 . The method of  claim 1 , wherein the first index entry comprises one or more known faces in the first reduced-resolution version. 
     
     
         4 . The method of  claim 1 , wherein the first index entry comprises one or more known human bodies based on body features. 
     
     
         5 . The method of  claim 1 , wherein the first index entry comprises metadata associated with the first reduced-resolution version, wherein the metadata is selected from the group consisting of: date or time when the first full-resolution image was created, location where the first full-resolution image was created, identification of a camera that took the first full-resolution image, and identification of camera attributes that took the first full-resolution image. 
     
     
         6 . The method of  claim 1 , wherein transmitting the first full-resolution image to the user is in response to a determination that the user is authorized to access the first full-resolution version. 
     
     
         7 . The method of  claim 1 , wherein performing semantic analysis on the respective reduced-resolution version comprises applying optical character recognition to extract one or more respective keywords from the respective reduced resolution version. 
     
     
         8 . A computer system for managing an image catalog, comprising:
 one or more processors;   memory; and   one or more programs stored in the memory and configured for execution by the one or more processors, the one or more programs comprising instructions for, at the computer system:
 receiving reduced-resolution versions of one or more images stored in an image database of an external service distinct from the one or more servers, and receiving a specified subject matter for the one or more images; 
 for each of the received reduced-resolution versions, creating a respective index entry in the image catalog, the respective index entry comprising one or more keywords extracted from the respective reduced-resolution version by performing semantic analysis on the respective reduced-resolution version using a convolutional neural network trained on images corresponding to the specified subject matter; 
 receiving a query from a user; 
 matching the query to a first index entry in the image catalog, the first index entry created for a first reduced-resolution version of a first full-resolution image stored in the image database of the external service; 
 retrieving the first full-resolution image from the image database of the external service; and 
 transmitting the first full-resolution image to the user. 
   
     
     
         9 . The computer system of  claim 8 , wherein the first index entry comprises a respective color palette for the first reduced-resolution version. 
     
     
         10 . The computer system of  claim 8 , wherein the first index entry comprises one or more known faces in the first reduced-resolution version. 
     
     
         11 . The computer system of  claim 8 , wherein the first index entry comprises one or more known human bodies based on body features. 
     
     
         12 . The computer system of  claim 8 , wherein the first index entry comprises metadata associated with the first reduced-resolution version, wherein the metadata is selected from the group consisting of: date or time when the first full-resolution image was created, location where the first full-resolution image was created, identification of a camera that took the first full-resolution image, and identification of camera attributes that took the first full-resolution image. 
     
     
         13 . The computer system of  claim 8 , wherein transmitting the first full-resolution image to the user is in response to a determination that the user is authorized to access the first full-resolution version. 
     
     
         14 . The computer system of  claim 8 , wherein performing semantic analysis on the respective reduced-resolution version comprises applying optical character recognition to extract one or more respective keywords from the respective reduced resolution version. 
     
     
         15 . A non-transitory computer readable storage medium storing one or more programs configured for execution by one or more processors of a computer system for managing an image catalog, the one or more programs comprising instructions for:
 receiving reduced-resolution versions of one or more images stored in an image database of an external service distinct from the one or more servers, and receiving a specified subject matter for the one or more images;   for each of the received reduced-resolution versions, creating a respective index entry in the image catalog, the respective index entry comprising one or more keywords extracted from the respective reduced-resolution version by performing semantic analysis on the respective reduced-resolution version using a convolutional neural network trained on images corresponding to the specified subject matter;   receiving a query from a user;   matching the query to a first index entry in the image catalog, the first index entry created for a first reduced-resolution version of a first full-resolution image stored in the image database of the external service;   retrieving the first full-resolution image from the image database of the external service; and   transmitting the first full-resolution image to the user.   
     
     
         16 . The computer readable storage medium of  claim 15 , wherein the first index entry comprises a respective color palette for the first reduced-resolution version. 
     
     
         17 . The computer readable storage medium of  claim 15 , wherein the first index entry comprises one or more known faces in the first reduced-resolution version. 
     
     
         18 . The computer readable storage medium of  claim 15 , wherein the first index entry comprises metadata associated with the first reduced-resolution version, wherein the metadata is selected from the group consisting of: date or time when the first full-resolution image was created, location where the first full-resolution image was created, identification of a camera that took the first full-resolution image, and identification of camera attributes that took the first full-resolution image. 
     
     
         19 . The computer readable storage medium of  claim 15 , wherein transmitting the first full-resolution image to the user is in response to a determination that the user is authorized to access the first full-resolution version. 
     
     
         20 . The computer readable storage medium of  claim 15 , wherein performing semantic analysis on the respective reduced-resolution version comprises applying optical character recognition to extract one or more respective keywords from the respective reduced resolution version.