tile2tile: Learning Game Filters for Platformer Style Transfer
This provides a tool for game designers and researchers to automatically adapt level designs across different platformer games, though it is incremental as it builds on existing style transfer and tile-based modeling techniques.
The authors tackled the problem of style transfer between tile-based platformer game levels by developing tile2tile, which uses models to translate level sketches into specific game styles, enabling cross-game transfers. They demonstrated this method on games like Super Mario Bros and Metroid, achieving functional style transfers as shown in their experiments.
We present tile2tile, an approach for style transfer between levels of tile-based platformer games. Our method involves training models that translate levels from a lower-resolution sketch representation based on tile affordances to the original tile representation for a given game. This enables these models, which we refer to as filters, to translate level sketches into the style of a specific game. Moreover, by converting a level of one game into sketch form and then translating the resulting sketch into the tiles of another game, we obtain a method of style transfer between two games. We use Markov random fields and autoencoders for learning the game filters and apply them to demonstrate style transfer between levels of Super Mario Bros, Kid Icarus, Mega Man and Metroid.