Research map / Probabilistic and scientific modeling / Differential equations and PDEs
EMMA: Extracting Multiple physical parameters from Multimodal Data
Farhat Shaikh, Ayan Banerjee, Sandeep Gupta
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 |
|---|---|---|---|
| add audio input to video (ablation) | EMMA · Rover (forced dynamics) | convergence epoch: 30 -> 5 | Improved |
| add audio input to video (ablation) | EMMA · Rover (forced dynamics) | training time (s): 122.3 -> 54.1 | Improved |
| add audio input to video (ablation) | EMMA · Rover mass estimate (GT 26.88 kg) | estimated mass (kg): 39.64 -> 27.79 | Improved |
| use LTC instead of Neural ODE (ablation) | EMMA · Rover (forcing input) | average parameter error: 34.1% -> 9.3% | Improved |
| use LTC instead of CT-GRU (ablation) | EMMA · Rover (forcing input) | average parameter error: 14.1% -> 9.3% | Improved |
| use LTC instead of Neural ODE (ablation) | EMMA · Pendulum (no forcing) | average parameter error: 15.0% -> 5.2% | Improved |
| use LTC instead of LSTM/GRU/Transformer (ablation) | EMMA · Pendulum (no forcing) | estimated L: 0.49 -> 0.49 (45cm) | No clear effect |
| use LTC instead of GRU (ablation) | EMMA · Rover (forced dynamics) | convergence epoch: 10 -> 5 | Improved |
| use LTC instead of Transformer (ablation) | EMMA · Rover (forced dynamics) | convergence epoch: 54 -> 5 | Improved |
| use LTC instead of GRU/LSTM/Transformer (ablation) | EMMA · Rover (forced dynamics) | mass estimate (kg): 39.50 -> 27.79 | Improved |
| replace YOLOv11 with optical flow tracking (ablation) | EMMA · Pendulum 45cm | estimated L (m): 0.50 -> 0.66 | Hurt |
| replace YOLOv11 with optical flow tracking (ablation) | EMMA · Rover | wheel radius r (m): 0.223 -> 0.202 | No clear effect |
Closest papers by meaning
- Learning Physics From Video: Unsupervised Physical Parameter Estimation for Continuous Dynamical Systems (CVPR 2025 · similarity 0.87)
- Integrating Multimodal Data for Joint Generative Modeling of Complex Dynamics (ICML 2024 · similarity 0.85)
- Extracting Governing Equations from Latent Dynamics via Multi-View Contrastive Learning (NeurIPS 2026 · similarity 0.84)
- Learning Stochastic Multiscale Models (NeurIPS 2025 · similarity 0.84)
- DynamicVerse: A Physically-Aware Multimodal Framework for 4D World Modeling (NeurIPS 2025 · similarity 0.84)
- WISA: World simulator assistant for physics-aware text-to-video generation (NeurIPS 2025 · similarity 0.84)
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.
