AI Paper Atlas

Research map / 3D vision / Gaussian splatting

WaterDescatterGS: Diverse Underwater 3D Scene Reconstruction Using Gaussian Splatting

shijun Zhou, Yuqing Xu, Xing Xie, Baojie Fan, Jiandong Tian

NeurIPS 2026 · 3D vision › Gaussian splatting · approach: Regularization and primitive design for 3DGS

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes WaterDescatterGS, a Gaussian splatting method for underwater 3D scene reconstruction that accounts for view-dependent backscatter to separate medium degradation from scene radiance.

Problem

Physics-based underwater 3D restoration and reconstruction usually model backscatter as a function of depth and attenuation, ignoring its view-dependence across camera poses, which hurts separating medium degradation from intrinsic radiance and reduces visual and geometric accuracy.

Approach

Models view-dependent underwater scattering within a Gaussian splatting framework so that backscatter is explicitly tied to camera pose, enabling separation of medium effects from scene appearance (details beyond the abstract not available).

Experimental setup

Builds on: 3D Gaussian Splatting

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy