Physics-Based Rigid Body Object Tracking and Friction Filtering From RGB-D Videos

TL;DR

This paper introduces a physics-based method for tracking 3D rigid objects and estimating their physical properties, like friction, from RGB-D videos using a differentiable physics model within an Extended Kalman Filter.

Contribution

It presents a novel real-to-sim approach that combines physics simulation with sensor data to track objects and estimate physical parameters in real-world scenarios.

Findings

Successfully tracks object trajectories and estimates friction coefficients.

Demonstrates effectiveness on synthetic and real-world datasets.

Provides publicly available benchmark datasets for future research.

Abstract

Physics-based understanding of object interactions from sensory observations is an essential capability in augmented reality and robotics. It enables to capture the properties of a scene for simulation and control. In this paper, we propose a novel approach for real-to-sim which tracks rigid objects in 3D from RGB-D images and infers physical properties of the objects. We use a differentiable physics simulation as state-transition model in an Extended Kalman Filter which can model contact and friction for arbitrary mesh-based shapes and in this way estimate physically plausible trajectories. We demonstrate that our approach can filter position, orientation, velocities, and concurrently can estimate the coefficient of friction of the objects. We analyze our approach on various sliding scenarios in synthetic image sequences of single objects and colliding objects. We also demonstrate and evaluate our approach on a real-world dataset. We make our novel benchmark datasets publicly available to foster future research in this novel problem setting and comparison with our method.