Modeling the Real World with High-Density Visual Particle Dynamics

TL;DR

This paper introduces HD-VPD, a high-density visual particle dynamics model utilizing novel Point Cloud Transformers to efficiently simulate complex physical scenes with over 100,000 particles, enabling advanced robotic motion analysis.

Contribution

The paper presents a new high-density world model with Point Cloud Transformers that significantly improves efficiency and quality in modeling physical dynamics of real-world scenes.

Findings

Interlacer dynamics is twice as fast as previous graph neural networks.

HD-VPD can handle 4x more particles with higher prediction quality.

Successfully models robotic tasks like box pushing and grasping.

Abstract

We present High-Density Visual Particle Dynamics (HD-VPD), a learned world model that can emulate the physical dynamics of real scenes by processing massive latent point clouds containing 100K+ particles. To enable efficiency at this scale, we introduce a novel family of Point Cloud Transformers (PCTs) called Interlacers leveraging intertwined linear-attention Performer layers and graph-based neighbour attention layers. We demonstrate the capabilities of HD-VPD by modeling the dynamics of high degree-of-freedom bi-manual robots with two RGB-D cameras. Compared to the previous graph neural network approach, our Interlacer dynamics is twice as fast with the same prediction quality, and can achieve higher quality using 4x as many particles. We illustrate how HD-VPD can evaluate motion plan quality with robotic box pushing and can grasping tasks. See videos and particle dynamics rendered by HD-VPD at https://sites.google.com/view/hd-vpd.