FluidGaussian：传播基于仿真的不确定性以实现功能智能的三维重建
FluidGaussian: Propagating Simulation-Based Uncertainty Toward Functionally-Intelligent 3D Reconstruction
将流体仿真诱导的不确定性作为主动学习的采样信号，指导多视角采集以提升几何与物理保真度，为物理感知的三维重建提供了可插拔的通用框架。
原文依据 We define a simulation-based uncertainty metric induced by fluid simulations and integrate it with active learning to prioritize views that improve both visual and physical fidelity.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论物理世界中的真实物体遵循物理定律，因此在与其他物理物体交互时行为合理。然而，当前从多视角二维图像对真实场景进行三维重建的方法主要优化视觉保真度，即使用光度损失训练并在图像或表示空间中推理不确定性。这种以外观为中心的视角忽略了物体间的接触与耦合，将功能关键区域（例如空气动力学或流体动力学表面）与装饰性区域混为一谈，并且即使加入物理正则项，重建的结构仍不够理想。所有这些都可能导致非物理且不合理的交互。为了解决这个问题，我们思考：三维重建如何能够超越视觉线索，感知真实世界的交互和底层物体功能？为了回答这个问题，我们提出FluidGaussian，一种即插即用的方法，将几何重建与普遍存在的流固耦合紧密耦合，以高粒度评估表面质量。我们定义了一个由流体仿真诱导的基于仿真的不确定性度量，并将其与主动学习相结合，以优先选择能够同时提升视觉和物理保真度的视角。在NeRF Synthetic（Blender）、Mip-NeRF 360和DrivAerNet++上的实证评估中，我们的FluidGaussian方法在流体仿真中实现了高达+8.6%的视觉峰值信噪比提升和-62.3%的速度散度降低。我们的代码可在https://github.com/delta-lab-ai/FluidGaussian获取。
英文原文摘要 1570 字符
Real objects that inhabit the physical world follow physical laws and thus behave plausibly during interaction with other physical objects. However, current methods that perform 3D reconstructions of real-world scenes from multi-view 2D images optimize primarily for visual fidelity, i.e., they train with photometric losses and reason about uncertainty in the image or representation space. This appearance-centric view overlooks body contacts and couplings, conflates function-critical regions (e.g., aerodynamic or hydrodynamic surfaces) with ornamentation, and reconstructs structures suboptimally, even when physical regularizers are added. All these can lead to unphysical and implausible interactions. To address this, we consider the question: How can 3D reconstruction become aware of real-world interactions and underlying object functionality, beyond visual cues? To answer this question, we propose FluidGaussian, a plug-and-play method that tightly couples geometry reconstruction with ubiquitous fluid-structure interactions to assess surface quality at high granularity. We define a simulation-based uncertainty metric induced by fluid simulations and integrate it with active learning to prioritize views that improve both visual and physical fidelity. In an empirical evaluation on NeRF Synthetic (Blender), Mip-NeRF 360, and DrivAerNet++, our FluidGaussian method yields up to +8.6% visual PSNR (Peak Signal-to-Noise Ratio) and -62.3% velocity divergence during fluid simulations. Our code is available in https://github.com/delta-lab-ai/FluidGaussian.