High-quality 4D reconstruction enables photorealistic and immersive rendering of the dynamic real world. However, unlike static scenes that can be fully captured with a single camera, high-quality dynamic scenes typically require dense arrays of tens or even hundreds of synchronized cameras. Dependence on such costly lab setups severely limits practical scalability. To this end, we propose a sparse-camera dynamic reconstruction framework that exploits abundant yet inconsistent generative observations. Our key innovation is the Spatio-Temporal Distortion Field, which provides a unified mechanism for modeling inconsistencies in generative observations across both spatial and temporal dimensions. Building on this, we develop a complete pipeline that enables 4D reconstruction from sparse and uncalibrated camera inputs. We evaluate our method on multi-camera dynamic scene benchmarks, achieving spatio-temporally consistent high-fidelity renderings and significantly outperforming existing approaches. Project page available at https://inspatio.github.io/sparse-cam4d/
@article{arxiv.2603.26481,
title = {SparseCam4D: Spatio-Temporally Consistent 4D Reconstruction from Sparse Cameras},
author = {Weihong Pan and Xiaoyu Zhang and Zhuang Zhang and Zhichao Ye and Nan Wang and Haomin Liu and Guofeng Zhang},
journal= {arXiv preprint arXiv:2603.26481},
year = {2026}
}
Comments
CVPR 2026. Project page: https://inspatio.github.io/sparse-cam4d/ and code: https://github.com/inspatio/sparse-cam4d