[2604.06358] GS-Surrogate: Deformable Gaussian Splatting for Parameter Space Exploration of Ensemble Simulations

Computer Science > Graphics arXiv:2604.06358 (cs) [Submitted on 7 Apr 2026] Title:GS-Surrogate: Deformable Gaussian Splatting for Parameter Space Exploration of Ensemble Simulations Authors:Ziwei Li, Rumali Perera, Angus Forbes, Ken Moreland, Dave Pugmire, Scott Klasky, Wei-Lun Chao, Han-Wei Shen View a PDF of the paper titled GS-Surrogate: Deformable Gaussian Splatting for Parameter Space Exploration of Ensemble Simulations, by Ziwei Li and 7 other authors View PDF HTML (experimental) Abstract:Exploring ensemble simulations is increasingly important across many scientific domains. However, supporting flexible post-hoc exploration remains challenging due to the trade-off between storing the expensive raw data and flexibly adjusting visualization settings. Existing visualization surrogate models have improved this workflow, but they either operate in image space without an explicit 3D representation or rely on neural radiance fields that are computationally expensive for interactive exploration and encode all parameter-driven variations within a single implicit field. In this work, we introduce GS-Surrogate, a deformable Gaussian Splatting-based visualization surrogate for parameter-space exploration. Our method first constructs a canonical Gaussian field as a base 3D representation and adapts it through sequential parameter-conditioned deformations. By separating simulation-related variations from visualization-specific changes, this explicit formulation enables efficient ...