logo

WorldGrow: Generating Infinite 3D World

1MoE Key Lab of Artificial Intelligence, School of Computer Science, SJTU
2Huawei Inc. 3Huazhong University of Science and Technology
*Equal contribution.Corresponding authors.

TL;DR: We propose WorldGrow — a generative method which creates infinite EXPLICIT 3D worlds, an alternative to the extensible, realistic, interactive world simulator.

Abstract

We tackle the challenge of generating the infinitely extendable 3D world — large, continuous environments with coherent geometry and realistic appearance. Existing methods face key challenges: 2D-lifting approaches suffer from geometric and appearance inconsistencies across views, 3D implicit representations are hard to scale up, and current 3D foundation models are mostly object-centric, limiting their applicability to scene-level generation. Our key insight is leveraging strong generation priors from pre-trained 3D models for structured scene block generation. To this end, we propose WorldGrow, a hierarchical framework for unbounded 3D scene synthesis. Our method features three core components: (1) a data curation pipeline that extracts high-quality scene blocks for training, making the 3D structured latent representations suitable for scene generation; (2) a 3D block inpainting mechanism that enables context-aware scene extension; and (3) a coarse-to-fine generation strategy that ensures both global layout plausibility and local geometric/textural fidelity. Evaluated on the large-scale 3D-FRONT dataset, WorldGrow achieves SOTA performance in geometry reconstruction, while uniquely supporting infinite scene generation with photorealistic and structurally consistent outputs. These results highlight its capability for constructing large-scale virtual environments and potential for building future world models.

Gallery

🏗️ Under construction: we are adding an 3DGS Viewer and a Mesh Viewer for interactive exploration (orbit / first-person, multi-view, material toggles).

A playable Web Demo / Game (walk-only navigation & planning) is also in the works. Stay tuned!

Pipeline

overview

Our goal is to generate infinite 3D scenes through modular, block-by-block synthesis. We begin by curating high-quality scene blocks and adapting SLAT to better model structured 3D context. A 3D block inpainting module enables spatially coherent extension, while a coarse-to-fine generation strategy ensures global layout plausibility and local detail fidelity. Together, these components allow WorldGrow to progressively construct photorealistic and structurally consistent 3D worlds with theoretically unbounded spatial extent.

Outdoor Worlds

Citation

@article{worldgrow2025,
  title   = {WorldGrow: Generating Infinite 3D World},
  author  = {Sikuang Li and Chen Yang and Jiemin Fang and Taoran Yi and Jia Lu and Jiazhong Cen and Lingxi Xie and Wei Shen and Qi Tian},
  journal = {arXiv preprint arXiv:2510.21682},
  year    = {2025}
}