用统一模型反推未知材料属性，实现更精准的物理仿真重演。
UniPhy: Learning a Unified Constitutive Model for Inverse Physics Simulation
- 基于共享潜空间的神经本构模型，跨材料泛化性强。
- 通过优化场景潜变量匹配观测运动，实现无需先验的逆向仿真。
- 支持弹性、黏土、沙粒、流体等多类材料，适合仿真与重建任务。
我们提出UniPhy，一种通用的潜变量条件神经本构模型，可编码多种材料的物理特性。推理时，UniPhy实现“逆向仿真”——通过可微分仿真优化场景特异的潜变量，以匹配实际观测。相比传统系统辨识方法，UniPhy无需用户指定材料类型；相较以往针对实例训练的神经模型，跨材料共享训练提升了估计的鲁棒性与准确性。模型在多样几何与材料（弹性、黏土、沙粒、流体（牛顿与非牛顿））的模拟轨迹上训练。给定未知材料属性的物体，UniPhy可通过潜变量优化推断其属性，并在新场景下重仿真。实验表明，相较于已有逆向仿真方法，UniPhy的推断结果能更准确地再现和重演复杂动态行为。
原文摘要 · Abstract (English)
We propose UniPhy, a common latent-conditioned neural constitutive model that can encode the physical properties of diverse materials. At inference UniPhy allows `inverse simulation' i.e. inferring material properties by optimizing the scene-specific latent to match the available observations via differentiable simulation. In contrast to existing methods that treat such inference as system identification, UniPhy does not rely on user-specified material type information. Compared to prior neural constitutive modeling approaches which learn instance specific networks, the shared training across materials improves both, robustness and accuracy of the estimates. We train UniPhy using simulated trajectories across diverse geometries and materials -- elastic, plasticine, sand, and fluids (Newtonian & non-Newtonian). At inference, given an object with unknown material properties, UniPhy can infer the material properties via latent optimization to match the motion observations, and can then allow re-simulating the object under diverse scenarios. We compare UniPhy against prior inverse simulation methods, and show that the inference from UniPhy enables more accurate replay and re-simulation under novel conditions.
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。