English

InfiniCity: Infinite-Scale City Synthesis

Computer Vision and Pattern Recognition 2023-08-16 v2 Artificial Intelligence Graphics Machine Learning

Abstract

Toward infinite-scale 3D city synthesis, we propose a novel framework, InfiniCity, which constructs and renders an unconstrainedly large and 3D-grounded environment from random noises. InfiniCity decomposes the seemingly impractical task into three feasible modules, taking advantage of both 2D and 3D data. First, an infinite-pixel image synthesis module generates arbitrary-scale 2D maps from the bird's-eye view. Next, an octree-based voxel completion module lifts the generated 2D map to 3D octrees. Finally, a voxel-based neural rendering module texturizes the voxels and renders 2D images. InfiniCity can thus synthesize arbitrary-scale and traversable 3D city environments, and allow flexible and interactive editing from users. We quantitatively and qualitatively demonstrate the efficacy of the proposed framework. Project page: https://hubert0527.github.io/infinicity/

Keywords

Cite

@article{arxiv.2301.09637,
  title  = {InfiniCity: Infinite-Scale City Synthesis},
  author = {Chieh Hubert Lin and Hsin-Ying Lee and Willi Menapace and Menglei Chai and Aliaksandr Siarohin and Ming-Hsuan Yang and Sergey Tulyakov},
  journal= {arXiv preprint arXiv:2301.09637},
  year   = {2023}
}
R2 v1 2026-06-28T08:18:05.807Z