US 12,106,432 B2
Machine learned resolution enhancement for virtual gaming environment
David Lupien St-Pierre, Amsterdam (NL); Noel Lopez-Gonzaga, Arnhem (NL); and Serge Vankeulen, Rotterdam (NL)
Assigned to PlayerUnknown Productions B.V., Amsterdam (NL)
Filed by PlayerUnknown Productions B.V., Amsterdam (NL)
Filed on Mar. 20, 2023, as Appl. No. 18/123,926.
Application 18/123,926 is a continuation of application No. 16/994,478, filed on Aug. 14, 2020, granted, now 11,607,611.
Claims priority of provisional application 62/946,960, filed on Dec. 11, 2019.
Prior Publication US 2023/0230314 A1, Jul. 20, 2023
Int. Cl. A63F 13/52 (2014.01); A63F 13/50 (2014.01); A63F 13/5378 (2014.01); A63F 13/69 (2014.01); A63F 13/80 (2014.01); G06F 18/214 (2023.01); G06N 3/045 (2023.01); G06T 17/00 (2006.01); G06T 17/05 (2011.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 20/13 (2022.01)
CPC G06T 17/00 (2013.01) [A63F 13/50 (2014.09); A63F 13/52 (2014.09); A63F 13/5378 (2014.09); A63F 13/69 (2014.09); A63F 13/80 (2014.09); G06F 18/2148 (2023.01); G06N 3/045 (2023.01); G06T 17/05 (2013.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 20/13 (2022.01); A63F 2300/8082 (2013.01)] 5 Claims
OG exemplary drawing
 
1. A method for generating a high-resolution basemap for use in providing graphics for use in a video game, the method comprising:
receiving elevation data from an image source;
processing elevation data from the image source into tiles of a determined size;
selecting an elevation dataset of images from tiles;
training a first machine learning agent based on the elevation dataset of images;
generating a low-resolution basemap image using the trained first machine learning agent;
dividing the low-resolution basemap image into a plurality of portions;
increasing the resolution of each of the plurality of portions to a higher resolution using a trained second machine learning agent;
deemphasizing border data of each the plurality of portions of the basemap image that have been resolution-enhanced; and
combining the resolution enhanced parts into a high-resolution basemap.