UW-SDF: Exploiting Hybrid Geometric Priors for Neural SDF Reconstruction from Underwater Multi-view Monocular Images

TL;DR

UW-SDF introduces a novel neural SDF-based framework utilizing hybrid geometric priors and a few-shot segmentation strategy to improve underwater 3D reconstruction quality and efficiency from multi-view images.

Contribution

The paper presents a new neural SDF reconstruction framework with hybrid geometric priors and a multi-view segmentation method for underwater environments.

Findings

Outperforms traditional underwater 3D reconstruction methods.

Enhances reconstruction quality and efficiency.

Enables rapid segmentation of unseen underwater objects.

Abstract

Due to the unique characteristics of underwater environments, accurate 3D reconstruction of underwater objects poses a challenging problem in tasks such as underwater exploration and mapping. Traditional methods that rely on multiple sensor data for 3D reconstruction are time-consuming and face challenges in data acquisition in underwater scenarios. We propose UW-SDF, a framework for reconstructing target objects from multi-view underwater images based on neural SDF. We introduce hybrid geometric priors to optimize the reconstruction process, markedly enhancing the quality and efficiency of neural SDF reconstruction. Additionally, to address the challenge of segmentation consistency in multi-view images, we propose a novel few-shot multi-view target segmentation strategy using the general-purpose segmentation model (SAM), enabling rapid automatic segmentation of unseen objects. Through extensive qualitative and quantitative experiments on diverse datasets, we demonstrate that our proposed method outperforms the traditional underwater 3D reconstruction method and other neural rendering approaches in the field of underwater 3D reconstruction.