Physically Embodied Gaussian Splatting: A Realtime Correctable World Model for Robotics

TL;DR

This paper introduces a real-time, dual Gaussian-Particle world model for robotics that combines physics simulation and visual correction to improve perception, planning, and interaction with the physical environment.

Contribution

It presents a novel dual representation integrating particles and 3D Gaussians for predictive simulation and visual correction in a unified, real-time system for robotics.

Findings

Runs at 30Hz with 3 cameras

Effective in 2D and 3D tracking tasks

Achieves high-quality photometric reconstruction

Abstract

For robots to robustly understand and interact with the physical world, it is highly beneficial to have a comprehensive representation - modelling geometry, physics, and visual observations - that informs perception, planning, and control algorithms. We propose a novel dual Gaussian-Particle representation that models the physical world while (i) enabling predictive simulation of future states and (ii) allowing online correction from visual observations in a dynamic world. Our representation comprises particles that capture the geometrical aspect of objects in the world and can be used alongside a particle-based physics system to anticipate physically plausible future states. Attached to these particles are 3D Gaussians that render images from any viewpoint through a splatting process thus capturing the visual state. By comparing the predicted and observed images, our approach generates visual forces that correct the particle positions while respecting known physical constraints. By integrating predictive physical modelling with continuous visually-derived corrections, our unified representation reasons about the present and future while synchronizing with reality. Our system runs in realtime at 30Hz using only 3 cameras. We validate our approach on 2D and 3D tracking tasks as well as photometric reconstruction quality. Videos are found at https://embodied-gaussians.github.io/.