Create a diffusion model for rendering video game frames. It takes N previous frames + a currently created frame and outputs a full rendered frame. The input frame should be created using ray tracing in real time with a much smaller resolution. For training, also a full resolution version should be created. The rendered output in the game could be even just material and lighting data without color rendering. For training, the fully rendered high resolution output is necessary though.
If you are interested in this topic or for more information please contact Prof. Helmut Hlavacs (helmut.hlavacs(at)univie.ac.at).