A neural network presented different variants of Dust2 from CS:GO
User named PittEnglishDept decided to experiment with Stable Diffusion neural network and ControlNet protocol to recreate Dust2 in different settings. He shared the results on reddit.
He chose the view of point A as the basis. The neural network created a cyberpunk map, a medieval village, a retro-futuristic city, abandoned ruins, an American settlement, and other locations. At the same time, the general features of Dust2 remained unchanged.
Previously, a CS:GO fan decided to recreate Dust2 in the style of Valorant. For this, he used Assets from the shooter from Riot Games, which he ported to the Unreal Engine 5.
Original Gangstar
April 6, 2023I would love to see a cyberpunk version of Dust2 as a playable map in CS:GO
Champion
April 6, 2023Looks so fresh 😀 why not
Choppa
April 6, 2023This is really cool! I love seeing how creative the gaming community can be when it comes to modding and recreating game content in different ways
Archie147
April 6, 2023Wow, these recreations look amazing! It’s fascinating how the general features of Dust2 remained unchanged despite the different settings