WorldCast enables scalable consistent views for multiplayer games
WorldCast: Distributed Multiplayer World Models
Computer Vision and Pattern Recognition
Summary
Playing multiplayer games means each player needs their own view of the game world that matches the others' views closely. Existing methods get slower as more players join because they generate everything together. The authors present WorldCast, where each player runs their own local system but shares player positions and scene information to keep everyone’s view consistent. This method can handle more players without slowing down and keeps the visual experience stable over long play sessions.
What this means in practice
- •For multiplayer game developers: Build multiplayer games that maintain consistent views across players without expensive centralized processing.$Commercial implications: Supports scalable, real-time multiplayer rendering making it possible to sell smoother multiplayer experiences with lower server costs.
- •For virtual reality platform engineers: Create distributed VR environments where multiple users see consistent shared scenes with low latency and high visual quality.
Authors
Ziyang Ye, Junchao Huang, Evelyn Zhang, Zhihao Xie, Ruicheng Zhang, Boyao Han, Litao Ban, Ziye Wang, Xinting Hu, Shaoshuai Shi, Zhuotao Tian, Li Jiang
Abstract
Multiplayer world models must generate independently controlled views with consistent representations of both players and their shared environment. Most existing approaches coordinate multiple players through joint multi-view generation, whose cost grows with each additional player. We present WorldCast, a distributed multiplayer world model in which each player runs a local client comprising a video generator and a state model. Using recorded player positions and map geometry during training, the state model estimates the player's position from generated video and control inputs. Clients exchange player states and project them into camera-aligned player state fields that guide where and how other players are rendered. Shared scene state enables clients to reuse one another's generated observations to maintain consistent scene appearance across views. Experiments on Counter-Strike 2 demonstrate WorldCast's consistency, real-time performance, and distributed scalability. The camera-aligned player state field improves player rendering rates by over an order of magnitude over joint-generation methods, while shared scene state improves visual consistency over whole rounds. Each client runs in real time and exchanges only player and scene states, enabling scalable multiplayer generation without a centralized computational bottleneck. Image quality remains stable over hour-long rollouts.