ChronoGS: Disentangling Invariants and Changes in Multi-Period Scenes

TL;DR

ChronoGS introduces a novel temporally modulated Gaussian representation for reconstructing multi-period scenes, effectively disentangling stable and changing components, and outperforms existing methods in reconstruction quality and consistency.

Contribution

The paper presents ChronoGS, a unified approach for multi-period scene reconstruction that handles long-term, discontinuous changes and introduces the ChronoScene dataset for benchmarking.

Findings

ChronoGS outperforms baselines in reconstruction quality.

ChronoGS achieves better temporal consistency.

The ChronoScene dataset captures diverse multi-period scenes.

Abstract

Multi-period image collections are common in real-world applications. Cities are re-scanned for mapping, construction sites are revisited for progress tracking, and natural regions are monitored for environmental change. Such data form multi-period scenes, where geometry and appearance evolve. Reconstructing such scenes is an important yet underexplored problem. Existing pipelines rely on incompatible assumptions: static and in-the-wild methods enforce a single geometry, while dynamic ones assume smooth motion, both failing under long-term, discontinuous changes. To solve this problem, we introduce ChronoGS, a temporally modulated Gaussian representation that reconstructs all periods within a unified anchor scaffold. It's also designed to disentangle stable and evolving components, achieving temporally consistent reconstruction of multi-period scenes. To catalyze relevant research, we release ChronoScene dataset, a benchmark of real and synthetic multi-period scenes, capturing geometric and appearance variation. Experiments demonstrate that ChronoGS consistently outperforms baselines in reconstruction quality and temporal consistency. Our code and the ChronoScene dataset are publicly available at https://github.com/ZhongtaoWang/ChronoGS.