LaGSplat: Learning Lagrangian Physics from Monocular Video
andrew_n_carr · x · 2026-08-31
LaGSplat introduces a method to learn Lagrangian dynamics of an object from just a few seconds of monocular video. By sharing a latent variable q to drive both Gaussian Splatting and equations of motion, the system allows for interactive simulation where external forces can be applied without measured force data in training.
More from Research
- LeVJEPA simplifies video self-supervised learning, cuts compute 5-20x — ylecun · 2026-08-30
- AI designs chip from spec to hardware in 2 weeks — rohanpaul_ai · 2026-08-30
- ForestDiffusion: XGBoost-based tabular data diffusion model favors CPU parallelization — jm_alexia · 2026-08-30
- Accio Open-Sources CommerceAgentBench, Claude Passes 52% — dr_cintas · 2026-08-30
- Elbow Method: Evaluating Optimal Cluster Count — mdancho84 · 2026-08-30
- Data Scientist Breaks Down K-means Clustering for Beginners — mdancho84 · 2026-08-30