# VoMP: Predicting Volumetric Mechanical Property Fields

arXiv: https://arxiv.org/abs/2510.22975 · HTML: https://mlsanity.com/paper/2510.22975/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Oct 2025.

## Built on by (most cited first, 13 of 13)

- [RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies](https://mlsanity.com/paper/2604.09860/) (2026)
- [PhysInOne: Visual Physics Learning and Reasoning in One Suite](https://mlsanity.com/paper/2604.09415/) (2026)
- [NeuROK: Generative 4D Neural Object Kinematics](https://mlsanity.com/paper/2605.30347/) (2026)
- [Axolotl3D: a Unified Framework for Faithful 3D Shape Completion](https://mlsanity.com/paper/2607.20660/) (2026)
- [FreeForm: Reduced-Order Deformable Simulation from Particle-Based Skinning Eigenmodes](https://mlsanity.com/paper/2605.29318/) (2026)
- [SiPhy: Single-Image Physical Property Reasoning](https://mlsanity.com/paper/2607.22355/) (2026)
- [SLAT-Phys: Fast Material Property Field Prediction from Structured 3D Latents](https://mlsanity.com/paper/2603.23973/) (2026)
- [EndoGSim: Physics-Aware 4D Dynamic Endoscopic Scene Simulations via MLLM-Guided Gaussian Splatting](https://mlsanity.com/paper/2605.16022/) (2026)
- [Gen2Physics: Grounding Generated 3D Meshes in Physics via Multi-View Material Decomposition](https://mlsanity.com/paper/2608.23869/) (2026)
- [LEIA: Learned Environment for Interactive Architected Materials](https://mlsanity.com/paper/2605.28368/) (2026)
- [TRON: Tracing Rays to Orchestrate a Neural Renderer for 3D Gaussian Reconstructions](https://mlsanity.com/paper/2606.11314/) (2026)
- [Vision Meets WiFi: Physics-Grounded Estimation of Volumetric Mechanical Properties](https://mlsanity.com/paper/2608.07726/) (2026)
- [PhysVGGT: Feed-Forward Dense Physical Property Estimation from A Single Image](https://mlsanity.com/paper/2609.18920/) (2026)

## Abstract (arXiv)

Physical simulation relies on spatially-varying mechanical properties, often laboriously hand-crafted. VoMP is a feed-forward method trained to predict Young's modulus (E), Poisson's ratio (nu), and density (ρ) throughout the volume of 3D objects, in any representation that can be rendered and voxelized. VoMP aggregates per-voxel multi-view features and passes them to our trained Geometry Transformer to predict per-voxel material latent codes. These latents reside on a manifold of physically plausible materials, which we learn from a real-world dataset, guaranteeing the validity of decoded per-voxel materials. To obtain object-level training data, we propose an annotation pipeline combining knowledge from segmented 3D datasets, material databases, and a vision-language model, along with a new benchmark. Experiments show that VoMP estimates accurate volumetric properties, far outperforming prior art in accuracy and speed.