Towards Spatio-Temporally Consistent 4D Reconstruction from Sparse Cameras

ICLR 2026 Conference Submission16080 Authors

Keywords: 4D Gaussian Splatting, Sparse View, Neural Rendering

Abstract: 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 scene benchmarks typically use dense arrays of approximately 20 synchronized cameras or more. The reliance 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.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 16080