Aesthetics-guided image enhancement

A computer-implemented method for generating images with enhanced aesthetics, the method comprising:

• selecting an image to aesthetically enhance;
• obtaining a segmentation map corresponding to the selected image;
• inputting the selected image and corresponding segmentation map into a trained neural network system based on a generative adversarial type architecture, the trained neural network system having a first neural network to convert the selected image into an enhanced aesthetic image utilizing the segmentation map;
• and
• outputting the enhanced aesthetic image generated using the first neural network of the trained neural network system.

A computing system comprising:

• one or more processors;
• and
• one or more non-transitory computer-readable storage media, coupled with the one or more processors, having instructions stored thereon, which, when executed by the one or more processors, cause the computing system to provide:
• means for training a neural network system, wherein the neural network system comprises a first neural network trained to enhance aesthetics of an image input into the first neural network by generating an enhanced image and a second neural network trained to evaluate realism of the enhanced image generated by the first neural network;
• and
• means for converting an image into an enhanced image using the trained first neural network of the trained neural network system.