OceanXL: Large-scale Underwater 3D Gaussian Splatting via Block Partitioning and Adaptive Pruning

2026-09-24

AI Summary

This study addresses the challenges of light attenuation, scattering, excessive memory overhead, and low optimization efficiency in large-scale underwater 3D reconstruction by proposing an efficient framework based on 3D Gaussian Splatting (3DGS). The method introduces a novel spatial chunking strategy and an adaptive pruning mechanism tailored for underwater scenes, effectively overcoming limited coverage bottlenecks to achieve compact, real-time rendering. Additionally, this work constructs the first large-scale underwater multi-environment dataset. Experiments across five real-world scenes demonstrate that the proposed framework exhibits superior scalability and an effective balance between efficiency and quality, yielding significantly smaller model sizes than existing methods while maintaining high-fidelity reconstruction.

Abstract

Underwater 3D reconstruction is critical for marine exploration, ecological monitoring, and subsea infrastructure inspection, yet remains challenging at large scale due to light attenuation, scattering, and limited capture coverage. While 3D Gaussian Splatting (3DGS) enables high-quality real-time rendering, its application to large underwater scenes is constrained by high memory consumption and inefficient optimization over extensive areas. We propose OceanXL, a fast and scalable 3DGS-based framework for large-scale underwater reconstruction. OceanXL adopts a divide-and-conquer strategy, partitioning scenes into spatially coherent blocks to enable efficient optimization while preserving global geometric consistency. We further introduce an adaptive pruning scheme tailored to underwater conditions that removes redundant primitives, producing compact representations without sacrificing visual fidelity. Together, these components improve training efficiency and rendering performance for large scenes. We also introduce a large-scale underwater dataset covering diverse marine environments. Experiments on five large-scale scenes demonstrate favorable scalability, compactness, and efficiency--quality trade-offs over large-scene baselines. Controlled comparisons on the small-scale SeaThru-NeRF dataset further show competitive reconstruction quality with substantially smaller model sizes than underwater-specific methods.

Problem

• Underwater 3D reconstruction
• Large-scale scenes
• 3D Gaussian Splatting
• Memory consumption
• Optimization efficiency

Innovation

• 3D Gaussian Splatting
• Underwater 3D Reconstruction
• Block Partitioning
• Adaptive Pruning
• Large-scale Scene