PREFERRED IMAGE RETRIEVAL
The present disclosure proposes computer-based systems and methods for determining, for an application, the most relevant subset of images across a number of sources of images by identifying images over multiple image providers, aggregating image collection over those images that are candidates for use, evaluating candidate images based on comparing image metadata against customizable image preference criteria, and producing a set of images to be selected either by a user or by a computer for use in the application.
1 . A system, comprising:
a memory that stores a plurality of images and computer-readable instructions, a separate attribute value for each of a plurality of attributes is associated with each corresponding image of the plurality of images; and
a computer processor that executes the computer-readable instructions to perform actions, the action including:
receiving, from a user, a plurality of image preference criteria for each of the plurality of attributes, each image preference criteria includes a separate potential attribute value;
determining, for each corresponding image of the plurality of images, a separate attribute score for each corresponding attribute of the plurality of attributes based on a comparison of the attribute value associated with the corresponding image for the corresponding attribute with the potential attribute value of each of the plurality of image preference criteria for the corresponding attribute;
combining the separate attribute scores to determine an image score for each of the plurality of images; and
providing, to the user, at least a ranked subset of the plurality of images based on the image score for each of the plurality of images.