WaterGS: Physically‐Based Imaging in Gaussian Splatting for Underwater Scene Reconstruction
- 11 October 2025
- journal article
- Published by Wiley in Computer Graphics Forum
- Vol. 44 (7)
Abstract
Reconstructing underwater object geometry from multi‐view images is a long‐standing challenge in computer graphics, primarily due to image degradation caused by underwater scattering, blur, and color shift. These degradations severely impair feature extraction and multi‐view consistency. Existing methods typically rely on pre‐trained image enhancement models as a preprocessing step, but often struggle with robustness under varying water conditions. To overcome these limitations, we propose WaterGS, a novel framework for underwater surface reconstruction that jointly recovers accurate 3D geometry and restores true object colors. The core of our approach lies in introducing a Physically‐Based imaging model into the rendering process of 2D Gaussian Splatting. This enables accurate separation of true object colors from water‐induced distortions, thereby facilitating more robust photometric alignment and denser geometric reconstruction across views. Building upon this improved photometric consistency, we further introduce a Gaussian bundle adjustment scheme guided by our physical model to jointly optimize camera poses and geometry, enhancing reconstruction accuracy. Extensive experiments on synthetic and real‐world datasets show that WaterGS achieves robust, high‐fidelity reconstruction directly from raw underwater images, outperforming prior approaches in both geometric accuracy and visual consistency.Keywords
Funding Information
- National Natural Science Foundation of China (61972379)
- Science and Technology Program of Suzhou (SYG202327)
This publication has 37 references indexed in Scilit:
- Method of 3D reconstruction of underwater concrete by laser line scanningOptics and Lasers in Engineering, 2024
- Refractive Pose RefinementInternational Journal of Computer Vision, 2023
- A Point Set Generation Network for 3D Object Reconstruction from a Single Image2017
- Pixelwise View Selection for Unstructured Multi-View Stereo2016
- High-Resolution Underwater Robotic Vision-Based Mapping and Three-Dimensional Reconstruction for ArchaeologyJournal of Field Robotics, 2016
- Large-Scale Data for Multiple-View StereopsisInternational Journal of Computer Vision, 2016
- True Color Correction of Autonomous Underwater Vehicle ImageryJournal of Field Robotics, 2015
- Perspective and Non-perspective Camera Models in Underwater Imaging – Overview and Error Analysis2012
- A Variational Framework for RetinexInternational Journal of Computer Vision, 2003
- Computer modeling and the design of optimal underwater imaging systemsIEEE Journal of Oceanic Engineering, 1990