Intel builds ultra-realistic GTAV using machine learning

GTA V A game that has been praised by many players around the world and is still very successful today, thanks to its unique identity on the one hand and its realism on the other. The game manages to shift the essence of Los Angeles, California to the city of San Andreas. However, it seems that this realism is not enough for the Intel Labs teams, they have developed a method to make the image in the game more realistic. Uses this method Mechanical learning And photosynthesis to achieve incredible results.

In the most complete video, Intel Labs teams explain all the steps needed to achieve this ultra-realistic result. In addition, they explain that this method is more stable than others used for the same purpose, including CUT and TSIT. In particular, it uses photos of real places, such as those taken in Germany from a car seat. With all the information in the photos and game data, the game adapts to create more realistic light, smooth road and more brightness on cars.

The only drawback to this simulation is that San Andreas loses its light and its California sun. It is true that the German climate is not the same as the west coast of the United States. However, we must admit that the result is shocking. Just like players can travel through a photo, you can almost count on the video tour provided Google Map. Not only does this look at what the Intel Labs project can do with technology, but it also makes one wonder if this kind of process can be implemented for greater realism in games in the future.