SurfPhase: 3D Interfacial Dynamics in Two-Phase Flows from Sparse Videos

Yue Gao*1     Hong-Xing (Koven) Yu*1     Sanghyeon Chang2     Qianxi Fu2     Bo Zhu3     Yoonjin Won2     Juan Carlos Niebles1     Jiajun Wu1
1Stanford University    2University of California, Irvine    3Georgia Institute of Technology
*Contributed Equally
Paper arXiv Code (Coming Soon) Dataset (Coming Soon)

Overview

SurfPhase overview figure

Novel View Video Synthesis

From a single input video, we reconstruct the fluid and show novel view synthesis results below. We show two novel views for each dataset example.

Interfacial Geometry Reconstruction

We visualize reconstructed interface meshes on real data and synthetic data.

Ablation Studies

Video Refinement

We ablate the video refinement stage by removing the diffusion-based novel-view video synthesis and the subsequent second-stage optimization, denoted as "SurfPhase w/o Refine".

Geometric Surface Constraint

We ablate the SDF formulation by removing the geometric surface constraint from the Gaussian surfel representation, denoted as "SurfPhase w/o SDF".

Dataset

We collect a dataset of high-speed pool boiling videos comprising 200 monocular videos and a synchronized dual-view video pair with physically calibrated metric scale, enabling training and evaluation for 3D interfacial reconstruction.

BibTex

@misc{gao2026surfphase,
    title={SurfPhase: 3D Interfacial Dynamics in Two-Phase Flows from Sparse Videos},
    author={Yue Gao and Hong-Xing Yu and Sanghyeon Chang and Qianxi Fu and Bo Zhu and Yoonjin Won and Juan Carlos Niebles and Jiajun Wu},
    year={2026},
    eprint={2602.11154},
    archivePrefix={arXiv},
    primaryClass={cs.CV},
    url={https://arxiv.org/abs/2602.11154},
}