Recent advances in novel-view synthesis can create the photo-realistic visualization of real-world environments from conventional camera captures. However, the everyday environment experiences frequent scene changes, which require dense observations, both spatially and temporally, that an ordinary setup cannot cover. We propose long-term Gaussian scene chronology from sparse-view updates, coined LTGS, an efficient scene representation that can embrace everyday changes from highly under-constrained casual captures. Given an incomplete and unstructured 3D Gaussian Splatting (3DGS) representation obtained from an initial set of input images, we robustly model the long-term chronology of the scene despite abrupt movements and subtle environmental variations. We construct objects as template Gaussians, which serve as structural, reusable priors for shared object tracks. Then, the object templates undergo a further refinement pipeline that modulates the priors to adapt to temporally varying environments given few-shot observations. Once trained, our framework is generalizable across multiple time steps through simple transformations, significantly enhancing the scalability for a temporal evolution of 3D environments. As existing datasets do not explicitly represent the long-term real-world changes with a sparse capture setup, we collect real-world datasets to evaluate the practicality of our pipeline. Experiments demonstrate that our framework achieves superior reconstruction quality compared to other baselines while enabling fast and light-weight updates.
Method Overview. We propose an integrated pipeline to update an initial reconstruction given the collection of post-change captures. Our pipeline first estimates the camera poses of the input capture and compares them against renderings of the initial reconstruction in the same view to detect object-level changes. We aggregate detected objects from multiple viewpoints and timestamps to create 3D Gaussian templates, and finally update the temporal scenes by compositing the templates at their respective states with the background.
Free-trajectory video demonstrating the initial reconstruction alongside the input images, followed by the complete temporally reconstructed scene.
Free-trajectory video of the scene showcasing the initial reconstruction, input images, and temporally reconstructed scene, including complex object articulation.
Qualitative comparisons of our method. We illustrate the results of our method using the CL-NeRF dataset and our dataset.
@article{kim2025ltgs,
title={LTGS: Long-Term Gaussian Scene Chronology From Sparse View Updates},
author={Kim, Minkwan and Lee, Seungmin and Kim, Junho and Kim, Young Min},
journal={arXiv preprint arXiv:2510.09881},
year={2025}
}