WAT3R: Feedforward Underwater 3D Reconstruction

2026-07-23

AI Summary

Underwater images suffer from visual degradation and inconsistent multi-view features due to light attenuation and backscattering, posing significant challenges for reliable feedforward 3D reconstruction. To address this, this work proposes WAT3R, a novel framework that introduces a lightweight neural adaptation module to explicitly model underwater imaging degradation as geometric constraints. Within a single feedforward pass, WAT3R jointly predicts pixel-aligned 3D point maps and camera poses by integrating a lightweight network, geometric optimization, and an end-to-end architecture. The method enables simultaneous depth and pose estimation from both monocular and multi-view inputs. Extensive experiments on the FLSea, SQUID, and USOD10K datasets demonstrate that WAT3R consistently outperforms state-of-the-art approaches in 3D reconstruction accuracy, depth estimation, and pose prediction.

Abstract

Reliable feedforward underwater 3D reconstruction remains challenging due to severe light attenuation and backscattering, which degrade visual quality and disrupt feature consistency across views, leading to inaccurate multi-view geometry. To address this issue, we propose WAT3R, a feed-forward framework for reconstructing 3D scenes directly from underwater images. By leveraging degradation adaptation as a geometry-constrained process, WAT3R integrates a lightweight neural adaptation module to flexibly account for these underwater imaging effects, thereby improving multi-view reconstruction quality. Implemented in a single forward pass, WAT3R directly and efficiently outputs pixel-aligned 3D point maps and camera poses from underwater videos, allowing a high-quality underwater 3D reconstruction. Experiments conducted on the FLSea, SQUID, and USOD10K datasets show that our method consistently outperforms state-of-the-art approaches on 3D reconstruction tasks, including multi-view/monocular depth estimation and camera pose estimation.

Problem

• underwater 3D reconstruction
• light attenuation
• backscattering
• multi-view geometry
• feature consistency

Innovation

• underwater 3D reconstruction
• feedforward framework
• degradation adaptation
• multi-view geometry
• neural adaptation module