Research map / Video / Video segmentation and tracking
Use the Force, Luke! Learning to Predict Physical Forces by Simulating Effects
Kiana Ehsani, Shubham Tulsiani, Saurabh Gupta, Ali Farhadi, Abhinav Gupta
Check 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 |
|---|---|---|---|
| Jointly optimize contact-point and keypoint-projection objectives instead of training the branches independently (ablation) | Force-and-contact prediction model · All objects test | Keypoint Error (px): 104.57 -> 99.38 | Improved |
| Jointly optimize contact-point and keypoint-projection objectives instead of training the branches independently (ablation) | Force-and-contact prediction model · All objects test | CP Error: 6.83e-2 -> 6.71e-2 | Improved |
| Use simulation gradients and keypoint-projection loss rather than regress pseudo-ground-truth forces (ablation) | Toy-airplane force prediction · Toy airplane test | Keypoint Error (px): 185.33 -> 105.60 | Improved |
| Regress pseudo-ground-truth forces instead of training through simulation (ablation) | Toy-airplane force prediction · Toy airplane train | Keypoint Error (px): 55.96 -> 59.86 | No clear effect |
| Provide ground-truth initial pose instead of predicting the initial state (ablation) | Toy-airplane force prediction · Toy airplane | Keypoint Error (px): 135.80 -> 128.55 | Improved |
| Jointly predict contact points and forces rather than predict forces on ground-truth contact points (ablation) | Force prediction · All objects test | Keypoint Error (px): 102.40 -> 99.38 | Improved |
Closest papers by meaning
- Force Prompting: Video Generation Models Can Learn And Generalize Physics-based Control Signals (NeurIPS 2025 · similarity 0.90)
- FEEL (Force-Enhanced Egocentric Learning): A Dataset for Physical Action Understanding (ECCV 2026 · similarity 0.89)
- Unsupervised Learning for Physical Interaction through Video Prediction (NeurIPS 2016 · similarity 0.89)
- Hand-Object Contact Force Estimation from Markerless Visual Tracking (TPAMI 2017 · similarity 0.88)
- PhysPT: Physics-aware Pretrained Transformer for Estimating Human Dynamics from Monocular Videos (CVPR 2024 · similarity 0.87)
- InterMimic: Towards Universal Whole-Body Control for Physics-Based Human-Object Interactions (CVPR 2025 · 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.
