Touch-GS: Visual-Tactile Supervised 3D Gaussian Splatting

TL;DR

This paper introduces a method to supervise 3D Gaussian Splatting scenes using optical tactile sensors, combining touch and vision data to improve 3D scene reconstruction, especially for challenging objects like transparent or reflective surfaces.

Contribution

The novel integration of optical tactile data with 3D Gaussian Splatting using a Gaussian Process Implicit Surface and a new variance weighted depth loss function.

Findings

Enhanced 3D scene reconstruction accuracy.

Improved results on opaque, reflective, and transparent objects.

Better performance with combined touch and vision data.

Abstract

In this work, we propose a novel method to supervise 3D Gaussian Splatting (3DGS) scenes using optical tactile sensors. Optical tactile sensors have become widespread in their use in robotics for manipulation and object representation; however, raw optical tactile sensor data is unsuitable to directly supervise a 3DGS scene. Our representation leverages a Gaussian Process Implicit Surface to implicitly represent the object, combining many touches into a unified representation with uncertainty. We merge this model with a monocular depth estimation network, which is aligned in a two stage process, coarsely aligning with a depth camera and then finely adjusting to match our touch data. For every training image, our method produces a corresponding fused depth and uncertainty map. Utilizing this additional information, we propose a new loss function, variance weighted depth supervised loss, for training the 3DGS scene model. We leverage the DenseTact optical tactile sensor and RealSense RGB-D camera to show that combining touch and vision in this manner leads to quantitatively and qualitatively better results than vision or touch alone in a few-view scene syntheses on opaque as well as on reflective and transparent objects. Please see our project page at http://armlabstanford.github.io/touch-gs