揭秘动态高斯点云的隐藏机制，提升重建稳定性与可复现性。
FreeTimeGS++: Secrets of Dynamic Gaussian Splatting and Their Principles
- 通过控制变量实验解析4DGS的关键隐性机制
- 发现时间分段由高斯持续时间自发形成，光照与运动可解耦
- 提出新方法实现更稳定、低波动的动态场景重建
近期在4D高斯点云（4DGS）上的进展实现了令人瞩目的动态场景重建效果。尽管这些方法表现出色，但其性能提升的具体因素仍缺乏系统研究，导致对底层原理的理解困难。本文对这些隐藏因素进行了全面分析，以更清晰地理解4DGS框架。我们首先建立一个受控基线FreeTimeGS_ours，通过形式化并复现当前最优方法FreeTimeGS的启发式策略。基于此框架，我们沿着4DGS的基本维度进行探究，识别出若干实用秘诀，包括由高斯持续时间驱动的涌现式时间分段，以及光照保真度与运动行为之间的解耦特性。基于上述洞察，我们提出FreeTimeGS++，一种基于门控消融、UFM引导初始化和颜色校正的原理性方法，显著提升了重建的稳定性与可复现性，有效降低运行间方差。
原文摘要 · Abstract (English)
Recent progress in 4D Gaussian Splatting (4DGS) has achieved impressive dynamic scene reconstruction results. While these methods demonstrate remarkable performance, the specific factors behind their gains remain underexplored, making a systematic understanding of the underlying principles challenging. In this paper, we perform a comprehensive analysis of these hidden factors to provide a clearer perspective on the 4DGS framework. We first establish a controlled baseline, FreeTimeGS_ours, by formalizing and reproducing the heuristics of the state-of-the-art FreeTimeGS. Using this framework, we examine 4DGS along its fundamental axes and identify practical secrets, including the emergent temporal partitioning driven by Gaussian durations and the decoupling between photometric fidelity and motion behavior. Based on these insights, we propose FreeTimeGS++, a principled method that employs gated marginalization, UFM-guided initialization, and color correction to improve stability and reproducibility. Our approach yields reproducible results with reduced run-to-run variance.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。