×

UTILIZING DEEP LEARNING FOR RATING AESTHETICS OF DIGITAL IMAGES

  • US 20170294010A1
  • Filed: 04/12/2016
  • Published: 10/12/2017
  • Est. Priority Date: 04/12/2016
  • Status: Active Grant
First Claim
Patent Images

1. In a digital medium environment for analyzing digital images, a method of estimating aesthetic quality of digital images using deep learning, the method comprising:

  • receiving a plurality of training images with user provided ratings;

    sampling the plurality of training images to identify pairs of training images with a predetermined characteristic relationship;

    training a neural network to output aesthetic quality scores for identified pairs of training images that maintain a relative difference of the user provided ratings for the identified pairs of training images; and

    utilizing the trained neural network to generate an aesthetic quality score for an input digital image.

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