Research map / 3D vision / Face and shape
VoMP: Predicting Volumetric Mechanical Property Fields
Rishit Dagli, Donglai Xiang, Vismay Modi, Charles Loop, Clement Fuji Tsang, Anka He Chen, Anita Hu, Gavriel State, David Levin I.W., Maria Shugrina
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use MatVAE latent space instead of a vanilla VAE (ablation) | MatVAE · MTD test set | reconstruction and distribution metrics | Improved |
| Predict material triplets directly instead of MatVAE latents (ablation) | Geometry Transformer · GVM test set | Young's modulus and Poisson's ratio error | Hurt |
| Use log-transform normalization of E and density instead of z-score or no log (ablation) | Geometry Transformer · GVM test set | prediction accuracy | Improved |
| Use l2 loss instead of l1 loss (ablation) | Geometry Transformer · GVM test set | all error metrics: l1 errors increase by a factor of 2-3x for most properties | Improved |
| Use DINOv2 or CLIP features instead of RGB colors (ablation) | Geometry Transformer (random init) · GVM test set | prediction accuracy | Improved |
| Initialize Geometry Transformer from TRELLIS weights (ablation) | Geometry Transformer with DINOv2/CLIP · GVM test set | prediction accuracy | Improved |
| Add total correlation penalty to MatVAE | MatVAE · MTD | latent disentanglement | Improved |
| Use VoMP instead of per-object optimization baselines | VoMP · Run-time on 53.9K-Gaussian objects | speedup: 5-100x | Improved |
Later papers that compared with this method
2 later accepted papers report VoMP as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Adaptive Volumetric Mechanical Property Fields Invariant to Resolution (ICML 2026 · better on 6 of 6 shared results)
- AutoPhyX: Automatic Text-Condition Physics Property Generation (ECCV 2026 · better on 2 of 2 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- OmniPhysGS: 3D Constitutive Gaussians for General Physics-Based Dynamics Generation (ICLR 2025 · similarity 0.86)
- PhysX-3D: Physical-Grounded 3D Asset Generation (NeurIPS 2025 · similarity 0.86)
- Physically Compatible 3D Object Modeling from a Single Image (NeurIPS 2024 · similarity 0.86)
- Vid2Sim: Generalizable, Video-based Reconstruction of Appearance, Geometry and Physics for Mesh-free Simulation (CVPR 2025 · similarity 0.85)
- MAGE : Single Image to Material-Aware 3D via the Multi-View G-Buffer Estimation Model (CVPR 2025 · similarity 0.85)
- PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics (CVPR 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
