Latent Intuitive Physics: Learning to Transfer Hidden Physics from A 3D Video
This addresses the challenge of transferring physics from visual data for applications in simulation and robotics, though it appears incremental as an extension of existing transfer learning and neural rendering approaches.
The paper tackles the problem of inferring hidden physical properties of fluids from a single 3D video and simulating them in new scenes, achieving strong performance in novel scene simulation, future prediction, and supervised particle simulation.
We introduce latent intuitive physics, a transfer learning framework for physics simulation that can infer hidden properties of fluids from a single 3D video and simulate the observed fluid in novel scenes. Our key insight is to use latent features drawn from a learnable prior distribution conditioned on the underlying particle states to capture the invisible and complex physical properties. To achieve this, we train a parametrized prior learner given visual observations to approximate the visual posterior of inverse graphics, and both the particle states and the visual posterior are obtained from a learned neural renderer. The converged prior learner is embedded in our probabilistic physics engine, allowing us to perform novel simulations on unseen geometries, boundaries, and dynamics without knowledge of the true physical parameters. We validate our model in three ways: (i) novel scene simulation with the learned visual-world physics, (ii) future prediction of the observed fluid dynamics, and (iii) supervised particle simulation. Our model demonstrates strong performance in all three tasks.