VAIR: Visuo-Acoustic Implicit Representations for Low-Cost, Multi-Modal Transparent Surface Reconstruction in Indoor Scenes

TL;DR

This paper introduces VAIR, a novel multi-modal implicit neural approach combining visual and acoustic data to accurately reconstruct transparent surfaces in indoor scenes, aiding mobile robot navigation.

Contribution

The paper presents a new fusion model using generative latent optimization for dense transparent surface reconstruction in indoor environments.

Findings

Significant improvement over state-of-the-art methods.

Effective multi-modal fusion of acoustic and visual data.

Successful reconstruction of transparent surfaces in indoor scenes.

Abstract

Mobile robots operating indoors must be prepared to navigate challenging scenes that contain transparent surfaces. This paper proposes a novel method for the fusion of acoustic and visual sensing modalities through implicit neural representations to enable dense reconstruction of transparent surfaces in indoor scenes. We propose a novel model that leverages generative latent optimization to learn an implicit representation of indoor scenes consisting of transparent surfaces. We demonstrate that we can query the implicit representation to enable volumetric rendering in image space or 3D geometry reconstruction (point clouds or mesh) with transparent surface prediction. We evaluate our method's effectiveness qualitatively and quantitatively on a new dataset collected using a custom, low-cost sensing platform featuring RGB-D cameras and ultrasonic sensors. Our method exhibits significant improvement over state-of-the-art for transparent surface reconstruction.