Systems, methods, and storage media for training a model for image evaluation

A system, comprising:

• one or more hardware processors configured by machine-readable instructions to:
• extract a first plurality of features from a plurality of first training images and a second plurality of features from a second training image;
• generate a model comprising a first image performance score for each of the plurality of first training images and a feature weight for each feature of the first plurality of features, the feature weight for each feature of the first plurality of features calculated based on an impact of a variation in the feature on first image performance scores of the plurality of first training images;
• and
• train the model by adjusting the impact of a variation of each of a first set of features of the first plurality of features that correspond to the second plurality of features from the second training image and adding a second image performance score for the second training image to the model,
• wherein the model is configured to output an image performance score for an image based on execution of the model using the image as input.

A method comprising:

• extracting, by a computing device, a first set of features from each of a plurality of training images;
• adding, by the computing device, an image performance score to the first set of features for each of the plurality of training images;
• generating, by the computing device, a model from the first set of features comprising the image performance score for each of the plurality of training images, the model comprising the image performance score for each of the plurality of training images according to a distribution of frequencies of features within the plurality of training images;
• executing, by the computing device, the model using a third set of features of a candidate image as input to calculate a predicted image performance score for the candidate image based on the distribution of frequencies of features;
• and
• generating, by the computing device, a record identifying the predicted image performance score for the candidate image.

A method, comprising:

• extracting, by a computing device, a first plurality of features from a plurality of first training images and a second plurality of features from a second training image;
• generating, by the computing device, a model comprising a first image performance score for each of the plurality of first training images and a weight for each feature of the first plurality of features, the weight for each feature of the first plurality of features calculated based on a frequency of the feature within the plurality of first training images;
• and
• training, by the computing device, the model by adjusting the frequency of each of a first set of features of the first plurality of features that correspond to the second plurality of features from the second training image and adding a second image performance score for the second training image to the model,
• wherein the model is configured to output an image performance score for an image based on execution of the model using the image as input.