Predicting the Physical Dynamics of Unseen 3D Objects
This addresses the problem of physical interaction prediction for robotics and virtual worlds, enabling generalization to novel objects, though it appears incremental as it builds on existing dynamics prediction methods.
The paper tackles the problem of predicting physical dynamics for unseen 3D objects after an impulsive force, focusing on changes in position, rotation, velocities, and stability. The result is a model that accurately generalizes to unseen object shapes and initial conditions, with experiments showing accurate predictions.
Machines that can predict the effect of physical interactions on the dynamics of previously unseen object instances are important for creating better robots and interactive virtual worlds. In this work, we focus on predicting the dynamics of 3D objects on a plane that have just been subjected to an impulsive force. In particular, we predict the changes in state - 3D position, rotation, velocities, and stability. Different from previous work, our approach can generalize dynamics predictions to object shapes and initial conditions that were unseen during training. Our method takes the 3D object's shape as a point cloud and its initial linear and angular velocities as input. We extract shape features and use a recurrent neural network to predict the full change in state at each time step. Our model can support training with data from both a physics engine or the real world. Experiments show that we can accurately predict the changes in state for unseen object geometries and initial conditions.