Neural compression and / or decompression of geodata

Abstract

System, comprising: one or more processors; and Memory that stores instructions that, in response to execution by the one or more processors, cause the system to perform operations that include: Obtaining an instance of geospatial data stored with a first set of bytes defining, for each of a plurality of geographic segments of a geographic coordinate system (GCS), a corresponding value for each of one or more variables; Projecting the instance of geodata into a plurality of images, each of the images being intended for a respective unique geographic area comprising a respective unique contiguous set of the geographic segments and being generated based on projecting the respective values ​​for the respective unique contiguous set of the geographic segments, and each of the images having an image dimension including one or more image channels and a height and a width for each of the image channels; generating a corresponding compressed representation for each of the images using a neural network encoder, wherein each of the corresponding compressed representations has a reduced dimension that is reduced relative to the image dimension, and wherein the corresponding compressed representations are stored together with a second set of bytes that is smaller than the first set of bytes with which the instance of geospatial data is stored; and Storing the corresponding compressed representations in one or more computer-readable media.