With the help of artificial intelligence machine learning, Intel Labs researchers Stephan Richter, Hassan Abu AlHaija, and Vladlen Koltun have created a photorealistic version of Grand Theft Auto 5’s open-world that makes anything we’ve seen yet from a gaming rig look like child’s play. Watch:
“Our method modifies images from the game to look more realistic. It is a convolutional network which produces images frame by frame and can be run at interactive rates,” the researchers say in the video, adding that the model used, called Cityscapes, contains “mostly images of German cities recorded with an automotive-grade camera.”
“Our method paves the roads with smoother asphalt, increases the glossiness of car paint, and greens the sun-dried hills of GTA’s California. The results are geometrically and semantically consistent with the input images and temporally stable.”
Translation? Our GTA V is a much better-looking game.
Now, this stuff is what we call ‘next-gen.’ Can you imagine if how good simulators like Gran Turismo and Project Cars will look once this kind of technology is incorporated into video games on a larger scale? How long do you figure it will be before this kind of machine learning makes its way into popular racing titles?
Top Gear Philippines is now on Quento! Click here to download the app and enjoy more articles and videos from Top Gear Philippines and your favorite websites.