Generating new image data via applying a computer vision model to two-dimensional arrays of pixels extracted from incoming streams of digitally encoded data packets

Abstract

A computing system is operable to decode a plurality of initial incoming streams of digitally encoded data packets to extract a plurality of digitally encoded image files from the plurality of initial incoming streams of digitally encoded data. A plurality of image correction data is generated via performing an image data processing function based on applying a computer vision model upon a plurality of digital image data extracted from the plurality of digitally encoded image files. A plurality of subsequent incoming streams of digitally encoded data packets are decoded to extract a plurality of additional digitally encoded image files from the plurality of subsequent incoming streams of digitally encoded data. A plurality of additional digital image data extracted from the plurality of additional digitally encoded image files are processed to generate a plurality of additional image correction data via re-performing the image data processing function.