发表机构
College of Computer Science, Nankai University; NKIARI; School of Automation, Southeast University; Advanced Ocean Institute of Southeast University, Nantong（南开大学计算机学院; 南开大学人工智能研究院（深圳）; 东南大学自动化学院; 东南大学南通先进海洋研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出两阶段框架3D-USE，通过MediumRBF、ATC与U-BAF实现水下场景级增强，在真实水下场景实验中提升了能见度与跨视图一致性且保留重建质量。
AI 中文摘要
水下三维重建可忠实地复现采集视角的色彩偏移与能见度损失，但物理逆变换可能在恢复的场景外观中留下估计误差。我们将水下场景级增强（Underwater Scene-level Enhancement, USE）定义为从退化的多视图水下观测中学习持久的、能见度增强的三维场景表示，以实现一致的增强渲染。实现USE既需要可靠的场景表示用于增强，也需要无配对增强三维数据的一致增强目标。因此，我们提出两阶段框架3D-USE：首先，介质径向基锚点表示（Medium Radial Basis Anchor Representation, MediumRBF）通过共享径向基锚点表示水体效应，并显式分解物体与介质的贡献，建立介质感知的高斯场景；基于该固定场景表示，外观转换一致性（Appearance Transition Consensus, ATC）将配对的二维水下图像增强（Underwater Image Enhancement, UIE）知识转换为场景全局与高斯局部目标，避免来自不一致增强视角的直接监督；随后，水下双边外观场（Underwater Bilateral Appearance Field, U-BAF）在高斯辐射度与介质外观中实现这些目标，推理时无需二维UIE模型即可直接渲染增强的新视角。在真实水下场景上的实验表明，该方法在保持重建质量的同时，提升了能见度与跨视图一致性。
英文摘要
Underwater 3D reconstruction faithfully reproduces the color shifts and visibility loss of captured views, while physical inversion may leave estimation errors in the recovered scene appearance. We formulate Underwater Scene-level Enhancement (USE) as learning a persistent, visibility-enhanced 3D scene representation from degraded multi-view underwater observations, enabling consistent enhanced rendering. Realizing USE requires both a reliable scene representation for enhancement and a consistent enhancement target without paired enhanced 3D data. Therefore, we present 3D-USE, a two-stage framework. First, the Medium Radial Basis Anchor Representation (MediumRBF) establishes a medium-aware Gaussian scene by representing water effects with shared radial-basis anchors and explicitly decomposing object and medium contributions. Based on this fixed scene representation, Appearance Transition Consensus (ATC) transfers paired 2D underwater image enhancement (UIE) knowledge into scene-global and Gaussian-local targets, avoiding direct supervision from inconsistent enhanced views. An Underwater Bilateral Appearance Field (U-BAF) then realizes these targets in Gaussian radiance and medium appearance. The scene directly renders enhanced novel views without a 2D UIE model at inference. Experiments on real underwater scenes show improved visibility and cross-view consistency while preserving reconstruction quality.
CommentsProject page: https://bilityniu.github.io/3D-USE/
