基于跨维高斯法向跃迁场的海底高保真场景重建

赵奚曼; 褚宣合; 陈瀚; 刘厶源

doi:10.11993/j.issn.2096-3920.2025-0167

基于跨维高斯法向跃迁场的海底高保真场景重建

doi: 10.11993/j.issn.2096-3920.2025-0167

大连海事大学轮机工程学院, 辽宁大连, 116026

基金项目: 国家自然科学基金项目资助(62301107); 辽宁省科学技术计划项目(2025JH2/101300066); 兴辽英才计划”青年拔尖人才项目(XLYC2403049).

详细信息

作者简介:
赵奚曼(2000-), 女, 硕士研究生, 主要研究方向为水下三维建模

通讯作者:
刘厶源(1990-), 男, 博士, 教授, 主要研究方向为水下智能感知与建模、机器学习.

中图分类号: TJ630; U663
计量
- 文章访问数: 4
- HTML全文浏览量: 5
- PDF下载量: 1
- 被引次数: 0
出版历程
- 收稿日期: 2025-12-16
- 修回日期: 2026-02-04
- 录用日期: 2026-02-09
- 网络出版日期: 2026-07-14

High-Fidelity Seafloor 3D Reconstruction Based on Cross-dimensional Gaussian Normal Transition Field

Marine Engineering College, Dalian Maritime University, Dalian 116026, China

摘要

摘要: 海洋科学勘测与水下环境探查等领域对海底高保真场景重建的需求日益提升, 三维高斯溅射技术作为一种先进的显式场景表示方法, 在海底环境重建与新视图合成任务中展现出显著应用前景。然而, 受水下成像介质模糊效应等影响, 重建结果常出现介质伪影与结构失真等缺陷, 严重制约其在实际水下复杂环境下的适用性。为解决上述问题, 文中提出一种基于跨维高斯法向跃迁场的海底高保真场景重建方法, 首先建立高斯基元的跨维映射与法向跃迁系统, 提升其对复杂结构的精细化几何建模能力; 其次提出高斯不透明度加权滤波模型, 抑制由介质模糊效应引发的重建伪影; 最后在多种水下场景的实验结果表明该方法具备高效处理复杂水下场景重建与新视图合成的能力。
- 海底场景重建 /
- 新视图合成 /
- 三维高斯溅射 /
- 水下介质成像 /
- 复杂结构建模
Abstract: The demand for high-fidelity scene reconstruction of the seafloor is growing in fields such as marine scientific surveying and underwater environmental exploration. As an advanced explicit scene representation method, 3D Gaussian Splatting shows significant application potential in scene reconstruction and novel view synthesis. However, influenced by factors such as blurring effects in underwater medium, the results often exhibit defects like medium artifacts and structural distortion, severely limiting their applicability in complex underwater environments. To address these challenges, we propose a high-fidelity seafloor scene reconstruction based on cross-dimensional Gaussian normal transition field. First, we establish a cross-dimensional and normal transition system for Gaussian primitives, enhancing its capability for detailed geometric modeling of complex structures. Second, we introduce a Gaussian opacity-weighted filtering model to suppress reconstruction artifacts caused by medium blurring effects. Finally, experimental results across multiple underwater scenes demonstrate our method's capability for efficient scene reconstruction and novel view synthesis in underwater environments.
- underwater scene reconstruction /
- novel view synthesis /
- 3d gaussian splatting /
- underwater medium imaging /
- complex structure modeling

HTML全文

图 1 跨维高斯法向跃迁场

Figure 1. Cross-dimensional Gaussian Normal Transition Field, GNTF

下载: 全尺寸图片幻灯片

图 2 高斯降维映射与法向跃迁

Figure 2. Gaussian cross-dimensional and normal transition

下载: 全尺寸图片幻灯片

图 3 水下高斯不透明度加权滤波

Figure 3. Gaussian opacity-weighted filtering

下载: 全尺寸图片幻灯片

图 4 海底真实场景

Figure 4. Seafloor real scenario

下载: 全尺寸图片幻灯片

图 5 不同方法对真实海底新视图合成的定性比较

Figure 5. Qualitative comparison of seafloor real scenario novel view synthesis using different methods

下载: 全尺寸图片幻灯片

图 6 不同方法对真实海底三维几何重建的定性比较

Figure 6. Qualitative comparison of seafloor real scenario 3D geometric reconstruction using different methods

下载: 全尺寸图片幻灯片

图 7 高斯降维映射与CNT消融实验对比图

Figure 7. Comparison of ablation experiments cross-dimensional and normal transition

下载: 全尺寸图片幻灯片

图 8 OWF消融实验对比图

Figure 8. Comparison of ablation experiments opacity-weighted filtering

下载: 全尺寸图片幻灯片

表 1 不透明度阈值调整分析

Table 1. Analysis of the opacity threshold adjustments

阈值	高斯数量(万)	FPS	PSNR	SSIM
0.005	128.4	18.3	36.91	0.982
0.05	87.5	23.4	37.22	0.989

下载: 导出CSV

表 2 不同方法下海底场景数据集新视图合成的定量比较

Table 2. Quantitative results of novel view synthesis for seafloor real scenario dataset

方法	PSNR	SSIM	LPIPS
NeRF	30.152	0.910	0.181
Instant NGP	31.091	0.929	0.169
Mip-NeRF	32.471	0.941	0.147
Tri-MipRF	32.976	0.950	0.129
3DGS	36.249	0.977	0.051
2DGS	34.590	0.969	0.088
Mip-Splatting	36.254	0.981	0.055
SuGaR	32.175	0.937	0.151
GNTF	37.224	0.989	0.036

下载: 导出CSV

表 3 三维几何重建定量对比

Table 3. Quantitative Comparison of 3D Geometric Reconstruction

方法	CD	运行时间/min
SfM+PSR	7.89	43.5
3DGS+Marching cubes	2.63	23.9
GNTF+Marching cubes	1.98	19.2

下载: 导出CSV

表 4 消融实验定量结果比较

Table 4. Quantitative comparison of ablation experiments

方法	PSNR	SSIM	LPIPS	CD
GNTF	37.224	0.989	0.036	1.93
GNTF w/o CNT	36.933	0.971	0.052	2.33
GNTF w/o OWF	36.897	0.984	0.049	2.01

下载: 导出CSV

参考文献(19)

[1]	Schonberger J L, Frahm J M. Structure-from-motion revisited[C]//Proceedings of the IEEE conference on computer vision and pattern recognition, 2016: 4104-4113.
[2]	Seitz S M, Curless B, Diebel J, et al. A comparison and evaluation of multi-view stereo reconstruction algorithms[C]//2006 IEEE computer society conference on computer vision and pattern recognition (CVPR'06) IEEE, 2006, 1: 519-528.
[3]	Dai P, Xu J, Xie W, et al. High-quality surface reconstruction using gaussian surfels[C]//ACM SIGGRAPH 2024 conference papers, 2024: 1-11.
[4]	Mildenhall B, Srinivasan P P, Tancik M, et al. Nerf: Representing scenes as neural radiance fields for view synthesis[J]. Communications of the ACM, 2021, 65(1): 99-106.
[5]	Barron J T, Mildenhall B, Tancik M, et al. Mip-nerf: A multiscale representation for anti-aliasing neural radiance fields[C]//Proceedings of the IEEE/CVF international conference on computer vision, 2021: 5855-5864.
[6]	Barron J T, Mildenhall B, Verbin D, et al. Mip-nerf 360: Unbounded anti-aliased neural radiance fields[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022: 5470-5479.
[7]	Hu W, Wang Y, Ma L, et al. Tri-miprf: Tri-mip representation for efficient anti-aliasing neural radiance fields[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023: 19774-19783.
[8]	Akkaynak D, Treibitz T. Sea-thru: A method for removing water from underwater images[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019: 1682-1691.
[9]	Levy D, Peleg A, Pearl N, et al. Seathru-nerf: Neural radiance fields in scattering media[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2023: 56-65.
[10]	Cong Y, Gu C, Zhang T, et al. Underwater robot sensing technology: A survey[J]. Fundamental Research, 2021, 1(3): 337-345. doi: 10.1016/j.fmre.2021.03.002
[11]	Kerbl B, Kopanas G, Leimkühler T, et al. 3D Gaussian splatting for real-time radiance field rendering[J]. ACM Transactions on Graphics, 2023, 42(4): 1-14.
[12]	Yu Z, Chen A, Huang B, et al. Mip-splatting: Alias-free 3d gaussian splatting[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2024: 19447-19456.
[13]	Huang B, Yu Z, Chen A, et al. 2d gaussian splatting for geometrically accurate radiance fields[C]//ACM SIGGRAPH 2024 conference papers, 2024: 1-11.
[14]	Hao Y, Yuan Y, Zhang H, et al. Underwater optical imaging: methods, applications and perspectives[J]. Remote Sensing, 2024, 16(20): 3773. doi: 10.3390/rs16203773
[15]	Chu X, Chen H, Zou D, et al. A multimodal optical dataset for underwater image enhancement, detection, segmentation, and reconstruction[J]. Scientific Data, 2025, 12(1): 1554. doi: 10.1038/s41597-025-05797-w
[16]	Yu Z, Sattler T, Geiger A. Gaussian opacity fields: Efficient adaptive surface reconstruction in unbounded scenes[J]. ACM Transactions on Graphics, 2024, 43(6): 1-13. doi: 10.1145/3687937
[17]	Lorensen W E, Cline H E. Marching cubes: A high resolution 3D surface construction algorithm[M]//Seminal graphics: pioneering efforts that shaped the field. 1998: 347-353.
[18]	Müller T, Evans A, Schied C, et al. Instant neural graphics primitives with a multiresolution hash encoding[J]. ACM transactions on graphics, 2022, 41(4): 1-15. doi: 10.1145/3528223.3530127
[19]	Guédon A, Lepetit V. Sugar: Surface-aligned gaussian splat-ting for efficient 3d mesh reconstruction and high-quality mesh rendering[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024: 5354-5363.