DLO-Splatting: Tracking Deformable Linear Objects Using 3D Gaussian Splatting

TL;DR

DLO-Splatting is a novel algorithm that estimates the 3D shape of deformable linear objects from multi-view images and gripper data, using shape modeling and Gaussian rendering to improve accuracy.

Contribution

It introduces a shape prediction-update filtering method with Gaussian splatting for deformable linear objects, advancing beyond existing vision-only approaches.

Findings

Promising results in knot tying scenarios

Effective shape estimation from multi-view RGB images

Improved alignment with visual observations

Abstract

This work presents DLO-Splatting, an algorithm for estimating the 3D shape of Deformable Linear Objects (DLOs) from multi-view RGB images and gripper state information through prediction-update filtering. The DLO-Splatting algorithm uses a position-based dynamics model with shape smoothness and rigidity dampening corrections to predict the object shape. Optimization with a 3D Gaussian Splatting-based rendering loss iteratively renders and refines the prediction to align it with the visual observations in the update step. Initial experiments demonstrate promising results in a knot tying scenario, which is challenging for existing vision-only methods.