Removing compression artifacts from digital images and videos utilizing generative machine-learning models

A system comprising:

• a memory component;
• and
• one or more processing devices coupled to the memory component, the one or more processing devices to perform operations comprising:
• extracting a compression ratio from a compressed digital image from metadata associated with the compressed digital image;
• extracting features from the compressed digital image;
• generating a combination of the compression ratio and the extracted features;
• generating an improved digital image by utilizing a generator neural network to remove compression artifacts from the compressed digital image by adjusting weighted parameters of a plurality of dilated attention residual neural network layers based on the combination of the compression ratio and the extracted features;
• and
• providing the improved digital image to a client device for display.

A method comprising:

• determining, utilizing one or more neural network layers, a compression ratio of a compressed digital image comprising a plurality of compression artifacts;
• extracting features from the compressed digital image;
• generating a combination of the compression ratio and the features extracted from the compressed digital image;
• a step for removing the plurality of compression artifacts from the compressed digital image based on the combination of the compression ratio and the features extracted from the compressed digital image utilizing a generator neural network;
• and
• providing a modified version of the compressed digital image to a client device for display.