Physics-as-Inverse-Graphics: Unsupervised Physical Parameter Estimation from Video
This addresses the problem of learning interpretable physical parameters without labeled data for researchers in computer vision and robotics, though it is incremental as it combines existing techniques.
The paper tackles unsupervised physical parameter estimation from video by integrating vision-as-inverse-graphics with differentiable physics, enabling long-term video prediction and vision-based control. It outperforms related unsupervised methods in future frame prediction for systems like ball-spring or 3-body gravitational systems and demonstrates data-efficient control for a pendulum.
We propose a model that is able to perform unsupervised physical parameter estimation of systems from video, where the differential equations governing the scene dynamics are known, but labeled states or objects are not available. Existing physical scene understanding methods require either object state supervision, or do not integrate with differentiable physics to learn interpretable system parameters and states. We address this problem through a physics-as-inverse-graphics approach that brings together vision-as-inverse-graphics and differentiable physics engines, enabling objects and explicit state and velocity representations to be discovered. This framework allows us to perform long term extrapolative video prediction, as well as vision-based model-predictive control. Our approach significantly outperforms related unsupervised methods in long-term future frame prediction of systems with interacting objects (such as ball-spring or 3-body gravitational systems), due to its ability to build dynamics into the model as an inductive bias. We further show the value of this tight vision-physics integration by demonstrating data-efficient learning of vision-actuated model-based control for a pendulum system. We also show that the controller's interpretability provides unique capabilities in goal-driven control and physical reasoning for zero-data adaptation.