用自适应体素预测3D物体的力学属性，精度与分辨率大幅提升。
Adaptive Volumetric Mechanical Property Fields Invariant to Resolution
- 基于稀疏自适应体素结构，自动为不同形状生成专属材料场。
- 相比之前方法，分辨率提升16³倍，测试时计算量更少。
- 适合需要高精度物理仿真的3D建模、游戏与工业设计场景。
精确的力学属性（杨氏模量 $E$、泊松比 $ν$、密度 $ρ$）对数字世界的真实物理仿真至关重要，但多数3D资产缺乏此类信息。我们提出AdaVoMP，一种可跨多种表示形式预测密集空间变化材料属性的方法，在分辨率、准确率和内存效率上均优于现有技术。其核心是稀疏自适应体素结构SAV，高效表示输入3D形状及输出材料场。将最准确的前序方法VoMP中的固定体素模型替换为新型稀疏Transformer编码器-解码器，实现对每种输入形状自回归生成唯一SAV，使分辨率达到前序方法的16³倍。实验表明，即使测试时计算开销更低，AdaVoMP仍能更准确估计体积属性，从而将高分辨率复杂3D物体转化为可直接用于仿真的资产，实现逼真的可变形模拟。
原文摘要 · Abstract (English)
Accurate mechanical properties (or materials) Young's modulus ($E$), Poisson's ratio ($ν$) and density ($ρ$) are essential for reliable physics simulation of digital worlds, but most 3D assets lack this information. We propose AdaVoMP, a method for predicting accurate dense spatially-varying ($E$, $ν$, $ρ$) for input 3D objects across representations, improving the resolution, accuracy, and memory efficiency over the state-of-the-art. The foundation of our technique is a sparse and adaptive voxel structure SAV that efficiently represents both the input 3D shape and the material field output. We replace the fixed-voxel model of the most accurate prior method, VoMP, with a novel sparse transformer encoder-decoder model that learns to generate a unique SAV autoregressively for every input shape to represent its materials, achieving a resolution $16^3\times$ higher than prior art. Experiments show that AdaVoMP estimates more accurate volumetric properties, even with lesser test-time compute than all prior art. This allows us to convert high-resolution complex 3D objects into simulation-ready assets, resulting in realistic deformable simulations.
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。