AnyGameRenderer: cross-game entity-consistent generative rendering
What happened
Ruicong Liu et al. propose AnyGameRenderer, a video-to-video generative rendering method, along with an entity-level benchmark. According to the paper, it can convert gray-box or flat-color game videos generated by AI agents into high-quality footage with characters, action details, and visual effects. The paper says the method performs this conversion while keeping the frame layout, camera, and the positions and timing of each entity consistent. The method is LoRA-fine-tuned on MiniMax-H3 ref2va, with about 10,000 video clip pairs for training, drawn from 38 agent-generated games, Unreal Engine procedural crowd data, and real game footage. The effects described come from the authors' own claims, and no independently verified results have been reported.
AI-written from the coverage · updated 2026-10-23
Coverage timeline
Follow the coverage to see the story from different sides.
- arXiv Game AIPickAnyGameRenderer: entity-consistent generative rendering across games and an entity-level benchmark
The paper proposes AnyGameRenderer, which converts gray-box or flat-color game videos generated by AI agents into high-quality footage with characters and effects, while keeping layout, camera, and the position and timing of each entity consistent. The method is LoRA-fine-tuned on MiniMax-H3 ref2va using about 10,000 video clip pairs from 38 agent-generated games, Unreal Engine procedural crowd data, and real game footage.