\textit{4DSurf}: High-Fidelity Dynamic Scene Surface Reconstruction

TL;DR

This paper introduces 4DSurf, a unified framework for high-fidelity dynamic scene surface reconstruction that handles large deformations and temporal inconsistencies without prior object specification.

Contribution

The paper proposes Gaussian deformations induced Signed Distance Function Flow Regularization and Overlapping Segment Partitioning to improve dynamic surface reconstruction over existing methods.

Findings

Outperforms state-of-the-art methods by 49% and 19% in Chamfer distance on two datasets.

Achieves superior temporal consistency under sparse-view settings.

Handles large deformations and does not require object type specification.

Abstract

This paper addresses the problem of dynamic scene surface reconstruction using Gaussian Splatting (GS), aiming to recover temporally consistent geometry. While existing GS-based dynamic surface reconstruction methods can yield superior reconstruction, they are typically limited to either a single object or objects with only small deformations, struggling to maintain temporally consistent surface reconstruction of large deformations over time. We propose ``\textit{4DSurf}'', a novel and unified framework for generic dynamic surface reconstruction that does not require specifying the number or types of objects in the scene, can handle large surface deformations and temporal inconsistency in reconstruction. The key innovation of our framework is the introduction of Gaussian deformations induced Signed Distance Function Flow Regularization that constrains the motion of Gaussians to align with the evolving surface. To handle large deformations, we introduce an Overlapping Segment Partitioning strategy that divides the sequence into overlapping segments with small deformations and incrementally passes geometric information across segments through the shared overlapping timestep. Experiments on two challenging dynamic scene datasets, Hi4D and CMU Panoptic, demonstrate that our method outperforms state-of-the-art surface reconstruction methods by 49\% and 19\% in Chamfer distance, respectively, and achieves superior temporal consistency under sparse-view settings.