发表机构
The Hong Kong University of Science and Technology; Beijing Institute of Technology（香港科技大学; 北京理工大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对水下环境光散射与动态物体导致的重建难题，提出首个介质感知前馈4D高斯溅射框架NemoSplat，设计可提示动态解缠器与介质感知高斯预测器，构建大规模水下数据集，实现了最优跟踪精度与高保真渲染。
AI 中文摘要
在非约束水下环境中重建逼真场景极具挑战，原因在于介质引发的严重光散射以及不可预测的动态物体。近期的前馈视觉基础模型在广义新视图合成与跟踪方面展现出卓越能力，但直接应用于水下视频时，光学衰减与运动干扰会严重破坏其特征聚合，导致跟踪与重建出现严重失败。为克服这些局限，我们提出NemoSplat，这是首个针对介质感知动态重建的前馈式4D高斯溅射框架，可直接从未校准的海洋视频中完成重建。除了能对相机位姿与密集场景深度进行稳健估计外，我们还设计了可提示动态解缠器，该解缠器利用学习到的动态概率与可选语义文本先验的置信度感知融合策略，有效分离大量瞬态实体。此外，为应对视觉退化，我们构建了介质感知高斯预测器，用于联合估计3D高斯的固有属性以及物理介质参数，通过单次前向传递渲染出原始场景外观。另外，我们引入了包含大量动态元素的大规模水下数据集，以助力训练与评估。在我们的数据集上开展的大量实验表明，NemoSplat实现了最先进的跟踪精度与高保真渲染效果。
英文摘要
Reconstructing photorealistic scenes in unconstrained underwater environments remains challenging due to severe media-induced light scattering and unpredictable dynamic objects. Recent feed-forward visual foundation models have demonstrated remarkable capabilities in generalized novel view synthesis and tracking. However, when directly applied to aquatic videos, optical attenuation and motion interference fatally corrupt their feature aggregation, leading to severe tracking and reconstruction failures. To overcome these limitations, we present NemoSplat, the first feed-forward 4D Gaussian Splatting framework tailored for media-aware dynamic reconstruction directly from uncalibrated marine videos. Beyond providing robust estimations of camera poses and dense scene depth, we devise a Promptable Dynamic Disentangler that utilizes a confidence-aware fusion strategy of learned dynamic probabilities and optional semantic text priors, effectively isolating massive transient entities. Furthermore, to counteract visual degradation, a Media-Aware Gaussian Predictor is formulated to jointly estimate intrinsic 3D Gaussian attributes alongside physical media parameters, rendering pristine scene appearance in a single forward pass. Additionally, we introduce a large-scale underwater dataset with massive dynamic elements to facilitate training and evaluation. Extensive experiments on our dataset demonstrate that NemoSplat achieves state-of-the-art tracking accuracy and high-fidelity rendering. Homepage: https://nemosplat.hkustvgd.com
Comments10 pages
