×

Iterative image search algorithm informed by continuous human-machine input feedback

  • US 10,467,267 B2
  • Filed: 08/28/2017
  • Issued: 11/05/2019
  • Est. Priority Date: 08/15/2014
  • Status: Active Grant
First Claim
Patent Images

1. A computer-implemented method for an improved iterative image search engine informed by human-machine input feedback, the method comprising the steps of:

  • initiating, using one or more computers, a search session in response to receiving an input from a human-machine interface device associated with a computer terminal having or operatively coupled to a video display device;

    retrieving, using at least one of the one or more computers, a set of digital images, each digital image associated with a set of tags from among a plurality of tags indicating one or more attributes or aspects of an object featured in the digital image;

    causing to be displayed, by the video display device, a next digital image from the set of digital images;

    repeating, a plurality of times until the search session ends, the following steps of;

    responsive to the causing to be displayed the next digital image, receiving, via the human-machine interface device, one of at least two input options, the at least two input options including a favorable indication of the next digital image and an unfavorable indication of the next digital image;

    analyzing, using at least one of the one or more computers, the set of tags associated with the next digital image to determine a subsequent set of tags that are to be associated with a subsequent digital image having at least a probable chance of being favorably indicated by a user, wherein the determining the subsequent set of tags is based on, responsive to the one of at least two input options selected via the human-machine interface device being the favorable indication, a weighting of tags to be included in the subsequent set of tags;

    responsive to the one of at least two input options selected via the human-machine interface device being the unfavorable indication, removing at least one tag of the set of tags associated with the next digital image from the plurality of tags for a remainder of the search session; and

    causing to be displayed, by the video display device, the subsequent digital image as the next digital image; and

    ending the search session in response to (a) receiving a further input via the human-machine interface device such that the next digital image displayed on the video display device is a final digital image of the search session, (b) the repeating occurring a predetermined number of times, (c) there remaining no further tags from the plurality of tags to select the subsequent digital image, or (d) there remaining no further digital images from the set of digital images to select the subsequent digital image,wherein each one of the at least two input options is received at least once during the search session.

View all claims
  • 3 Assignments
Timeline View
Assignment View
    ×
    ×