Wind on Trees: Testing Physical Grounding in Dynamic 4D Gaussian Splatting

2026-09-15

AI Summary

研究通过使用物理参数化的变形先验替代直接学习的变形场，以解决单目重建风驱动植被时的欠约束问题。

Abstract

Monocular reconstruction of wind-driven vegetation is severely underconstrained: motion along the viewing direction is largely unobservable, a moving canopy offers few reliable correspondences, and nearly the entire scene is dynamic, providing little static reference. Directly-learned deformation fields in 4D Gaussian Splatting therefore optimize photometric consistency rather than recover the motion that produced it. We replace that field with a physically parameterized deformation prior: one damped harmonic oscillator per rigid part, driven by the observed wind and integrated by differentiable RK4, supervised photometrically alone. To test whether such a prior is physically grounded rather than merely well fit, we build a controlled synthetic testbed of three procedurally generated trees spanning an order of magnitude in skeleton complexity, whose per-part natural frequency follows from its own geometry and whose damping ratio is a fixed constant, both held out of training. On it, we measure held-out views, temporal extrapolation, zero-shot transfer to unseen wind speeds, and recovery of the physical parameters themselves. The prior costs appearance fidelity on in-distribution views and extrapolates markedly better outside the training window and the training wind, while parameter recovery is far weaker than it first appears: frequency recovery survives an untrained null control on only the sparsest of the three trees, and damping is not recovered at all.

Problem

• Monocular Reconstruction
• Wind-Driven Vegetation
• Underconstrained
• Photometric Consistency
• Deformation Prior

Innovation

• physically parameterized deformation prior
• differentiable RK4 integrator
• synthetic testbed
• monocular reconstruction
• wind-driven vegetation