Research map / 3D vision / Gaussian splatting
AquaSplatting: A Hybrid 3D Representation for Robust Underwater Scene Reconstruction via Dual-Branch Rendering
Jiangbei Hu, Haobo Wang, Baixin Xu, Nan Ding, Zhimao Lu, Na Lei, Ying He
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add opacity-gated hybrid rendering vs direct sum of branches (ablation) | AquaSplatting · U-IW (Underwater in the Wild) | PSNR: 28.01 -> 28.25 | Improved |
| use both branches instead of geometry branch only (ablation) | AquaSplatting · U-IW (Underwater in the Wild) | PSNR: 27.98 -> 28.25 | Improved |
| use geometry branch instead of medium branch only (ablation) | AquaSplatting · U-IW (Underwater in the Wild) | PSNR: 25.53 -> 27.98 | Improved |
| add medium-aware density loss Ld (ablation) | AquaSplatting · SeaThru-NeRF loss ablation | PSNR: 29.82 -> 29.98 | Improved |
| add exposure loss (ablation) | AquaSplatting · SeaThru-NeRF loss ablation | PSNR: 29.98 -> 30.18 | Improved |
| add remaining losses (depth etc.) (ablation) | AquaSplatting · SeaThru-NeRF loss ablation | PSNR: 30.18 -> 30.32 | Improved |
| add engagement-based pruning (EBP) (ablation) | AquaSplatting · SeaThru-NeRF dataset | Gaussian count: up to 3.2x fewer Gaussians | Improved |
| add semantic mask supervision from Grounded SAM (ablation) | AquaSplatting · SeaThru-NeRF dataset | quality | No clear effect |
| use AquaSplatting vs Watersplatting | Watersplatting · U-IW (Underwater in the Wild) | PSNR: 25.35 -> 28.25 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- HoGS — HoGS: Unified Near and Far Object Reconstruction via Homogeneous Gaussian Splatting (CVPR 2025) · better on 2 of 3 shared results
- SeaThru-NeRF — SeaThru-NeRF: Neural Radiance Fields in Scattering Media (CVPR 2023) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- OceanSplat: Object-aware Gaussian Splatting with Trinocular View Consistency for Underwater Scene Reconstruction (AAAI 2026 · similarity 0.93)
- WaterDescatterGS: Diverse Underwater 3D Scene Reconstruction Using Gaussian Splatting (NeurIPS 2026 · similarity 0.93)
- Plenodium: Underwater 3D Scene Reconstruction with Plenoptic Medium Representation (NeurIPS 2025 · similarity 0.89)
- Atlantis: Enabling Underwater Depth Estimation with Stable Diffusion (CVPR 2024 · similarity 0.88)
- RestorGS: Depth-aware Gaussian Splatting for Efficient 3D Scene Restoration (CVPR 2025 · similarity 0.88)
- Synergistic Multiscale Detail Refinement via Intrinsic Supervision for Underwater Image Enhancement (AAAI 2024 · similarity 0.87)
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.
