Generation of training data for improved super-resolution models or downscaler networks

A method, implemented in a computing environment comprising at least one hardware processor and at least one memory coupled to the at least one hardware processor, the method comprising:

• receiving high-resolution images generated by a first image generator;
• processing the high-resolution images with a downscaler network to generate corresponding low-resolution images;
• generating a first set of a plurality of image pairs, each image pair comprising a high-resolution image from the first image generator and a corresponding low-resolution image produced by the downscaler network;
• and
• training a super-resolution model using the first set of image pairs.

One or more computer-readable storage media comprising:

• computer-executable instructions that, when executed by a computing system comprising at least one hardware processor and at least one memory coupled to the at least one hardware processor, cause the computing system to receive input low-resolution images from a first image generator executing on the computing system;
• computer-executable instructions that, when executed by the computing system, cause the computing system to process the input low-resolution images with a trained super-resolution model to generate enhanced images;
• computer-executable instructions that, when executed by the computing system, cause the computing system to provide the enhanced images for rendering on a display device of the computing system;
• and
• wherein the trained super-resolution model was generated using training data comprising high-resolution images and corresponding low-resolution images, the training data having been obtained from a training process that included:
• generating high-resolution images and corresponding low-resolution images using a second image generator configured with a rendering package;
• training a downscaler network using a first set of a plurality of image pairs of the high-resolution and low-resolution images;
• generating additional high-resolution images with the second image generator and processing the additional high-resolution images with the downscaler network to produce additional low-resolution images;
• generating a second set of a plurality of image pairs from the additional high-resolution images and corresponding additional low-resolution images produced by the downscaler network;
• and
• training the super-resolution model using the second set of image pairs.