OceanSplat: Object-aware Gaussian Splatting with Trinocular View Consistency for Underwater Scene Reconstruction

TL;DR

OceanSplat is a novel 3D Gaussian Splatting method that uses trinocular view consistency and synthetic depth priors to improve underwater scene reconstruction, reducing artifacts caused by scattering media.

Contribution

It introduces a trinocular setup and depth-aware regularization techniques to enhance 3D Gaussian Splatting for underwater scene reconstruction.

Findings

Outperforms existing methods in real-world underwater scenes

Reduces floating artifacts and improves scene fidelity

Effective in both simulated and real underwater environments

Abstract

We introduce OceanSplat, a novel 3D Gaussian Splatting-based approach for high-fidelity underwater scene reconstruction. To overcome multi-view inconsistencies caused by scattering media, we design a trinocular setup for each camera pose by rendering from horizontally and vertically translated virtual viewpoints, enforcing view consistency to facilitate spatial optimization of 3D Gaussians. Furthermore, we derive synthetic epipolar depth priors from the virtual viewpoints, which serve as self-supervised depth regularizers to compensate for the limited geometric cues in degraded underwater scenes. We also propose a depth-aware alpha adjustment that modulates the opacity of 3D Gaussians during early training based on their depth along the viewing direction, deterring the formation of medium-induced primitives. Our approach promotes the disentanglement of 3D Gaussians from the scattering medium through effective geometric constraints, enabling accurate representation of scene structure and significantly reducing floating artifacts. Experiments on real-world underwater and simulated scenes demonstrate that OceanSplat substantially outperforms existing methods for both scene reconstruction and restoration in scattering media.