浏览全部资源
扫码关注微信
1.成都信息工程大学 计算机学院
2.中国民用航空飞行学院 民航飞行技术与飞行安全科研基地,四川 成都 610225
朱烨(1975-),女,博士,成都信息工程大学计算机学院副教授、硕士生导师,研究方向为图像处理及应用
郑恒杰(2000-),男,成都信息工程大学计算机学院硕士研究生,研究方向为计算机图形学
陈闯。
戈文一(1986-),男,成都信息工程大学计算机学院助理研究教授,研究方向为遥感图像增强
谭诗瀚(1986-),男,成都信息工程大学计算机学院助理研究教授,研究方向为计算机图形学、虚拟现实开发的实时渲染系统飞行模拟
袁霞(1987-),女,成都信息工程大学计算机学院助理研究教授,研究方向为图形与图像处理
王盛杰(1989-),男,中国民用航空飞行学院民航飞行技术与飞行安全科研基地助理研究员,研究方向为光学工程、机器视觉。
收稿:2025-02-27,
纸质出版:2026-04-15
移动端阅览
朱烨,郑恒杰,陈闯等.WaterGs:一种基于3D高斯的动态水体生成方法[J].软件导刊,2026,25(04):148-154.
ZHU Ye,ZHENG Hengjie,CHEN Chuang,et al.WaterGs:A Dynamic Water Generation Method Based on 3D Gaussian[J].Software Guide,2026,25(04):148-154.
朱烨,郑恒杰,陈闯等.WaterGs:一种基于3D高斯的动态水体生成方法[J].软件导刊,2026,25(04):148-154. DOI: 10.11907/rjdk.242041.
ZHU Ye,ZHENG Hengjie,CHEN Chuang,et al.WaterGs:A Dynamic Water Generation Method Based on 3D Gaussian[J].Software Guide,2026,25(04):148-154. DOI: 10.11907/rjdk.242041.
动态水体生成方法是计算机图形学的主要研究任务之一。传统的水体生成方式并不能根据二维照片生成动态的三维水体,导致水体生成过程繁琐且缺乏真实性。为解决上述问题,提出一种基于3D高斯的水体生成方法WaterGs。首先,在静态重建过程中引入三维线性插值,通过填补缺失的数据点以保证静态水体模型表面的连续性和准确性;其次,根据不透明度阈值确定水体内外边界,对水体模型进行内部粒子填充,以提升动态重建效果;最后,将求解得到的水体运动轨迹、速度通过物理嵌入技术赋予水体模型,实现了水体动态生成。实验结果表明,WaterGs的PSNR、SSIM指标为30.92和0.859,与基准算法相比分别提升了7.58%和15.2%,并且实现了静态水体的动态化,显示出了其在水体生成中的优势。
Dynamic water generation is a key task in computer graphics. Traditional methods struggle to generate dynamic 3D water from 2D images, leading to a complicated and unrealistic process. To solve this, we propose a 3D Gaussian-based water generation method called WaterGs. In the static reconstruction phase, 3D linear interpolation is applied to fill missing data points, ensuring the continuity and accuracy of the water surface. Next, the internal and external boundaries of the water model are determined using opacity thresholds, and particles are added to enhance dynamic reconstruction. Finally, the water motion trajectories and velocities are embedded into the model using physical embedding techniques to achieve dynamic water generation. The experimental results show that WaterGs achieves PSNR and SSIM of 30.92 and 0.859, improving by 7.58% and 15.2% over the baseline, respectively. It also enables dynamic rendering of static water, highlighting its advantages in water generation.
LI Y L , ZHU G B , WANG R J , et al . The exploration of talent training model for digital twin practical skills under the integration of industry and education [J]. Computer Education , 2023 ( S1 ): 133 - 136 .
李彝利 , 朱国斌 , 王瑞锦 , 等 . 产教融合下的数字孪生实战人才培养模式探索 [J]. 计算机教育 , 2023 ( S1 ): 133 - 136 .
BEN M , PRATUL P , MATTHEW T , et al . Nerf: representing scenes as neural radiance fields for view synthesis [J]. Communications of the ACM , 2021 , 65 ( 1 ): 405 - 421 . doi: 10.1145/3503250 http://dx.doi.org/10.1145/3503250
KNAPITSCH A , PARK J , ZHOU Q Y , et al . Tanks and temples: benchmarking large-scale scene reconstruction [J]. ACM Transactions on Graphics , 2017 , 36 ( 4 ): 1 - 13 . doi: 10.1145/3072959.3073599 http://dx.doi.org/10.1145/3072959.3073599
FOURNIER A , REEVES W . A simple model of ocean waves [J]. ACM SIGGRAPH Computer Graphics , 1986 , 20 ( 4 ): 75 - 84 . doi: 10.1145/15886.15894 http://dx.doi.org/10.1145/15886.15894
TESSENDORF J . Simulating ocean water [J]. SIGGRAPH , 2001 , 1 ( 2 ): 5 .
JENSEN L , ROBERT G . Deep-water animation and rendering [C]// San Francisco: Game Developer’s Conference (Gamasutra) , 2001 .
TOMAS S , SODERSTROM A , JOHANSSON J , et al . Wave curves: simulating lagrangian water waves on dynamically deforming surfaces [J]. ACM Transactions on Graphics , 2020 , 39 ( 4 ): 1 - 11 . doi: 10.1145/3386569.3392466 http://dx.doi.org/10.1145/3386569.3392466
YUKSEL C , HOUSE H , KEYSER J , et al . Wave particles [J]. ACM Transactions on Graphics , 2007 , 26 ( 3 ): 99 -es. doi: 10.1145/1275808.1276501 http://dx.doi.org/10.1145/1275808.1276501
LI Y , LIN Z H , FORSYTH D , et al . Climatenerf: extreme weather synthesis in neural radiance field [C]// Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023 : 3204 - 3215 . doi: 10.1109/iccv51070.2023.00299 http://dx.doi.org/10.1109/iccv51070.2023.00299
PUMAROLA A , CORONA E . D-NeRF: neural radiance fields for dynamic scenes [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2021 : 10318 - 10327 . doi: 10.1109/cvpr46437.2021.01018 http://dx.doi.org/10.1109/cvpr46437.2021.01018
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 . doi: 10.1145/3592433 http://dx.doi.org/10.1145/3592433
WU G , YI T , FANG J , et al . 4D Gaussian splatting for real-time dynamic scene rendering [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2024 : 20310 - 20320 . doi: 10.1109/cvpr52733.2024.01920 http://dx.doi.org/10.1109/cvpr52733.2024.01920
LUITEN J , KOPANAS G , LEIBE B , et al . Dynamic 3D Gaussians: tracking by persistent dynamic view synthesis [C]// International Conference on 3D Vision (3DV) , 2024 : 800 - 809 . doi: 10.1109/3dv62453.2024.00044 http://dx.doi.org/10.1109/3dv62453.2024.00044
YANG Z , YANG H , PAN Z , et al . Real-time photorealistic dynamic scene representation and rendering with 4d Gaussian splatting[DB/OL] . https://arxiv.org/abs/2310.10642 https://arxiv.org/abs/2310.10642 .
YANG Z , GAO X , ZHOU W , et al . Deformable 3D Gaussians for high-fidelity monocular dynamic scenereconstruction [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2024 : 20331 - 20341 .
LIANG Y , KHAN N , LI Z , et al . Gaufre: Gaussian deformation fields for real-time dynamic novel view synthesis[DB/OL] . https://arxiv.org/abs/2312.11458 https://arxiv.org/abs/2312.11458 . doi: 10.1109/wacv61041.2025.00262 http://dx.doi.org/10.1109/wacv61041.2025.00262
LI Z , CHEN Z , LI Z , et al . Spacetime Gaussian feature splatting for real-time dynamic view synthesis [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2024 : 8508 - 8520 . doi: 10.1109/cvpr52733.2024.00813 http://dx.doi.org/10.1109/cvpr52733.2024.00813
XIE T , ZONG Z , QIU Y , et al . PhysGaussian: physics-integrated 3D Gaussians for generative dynamics [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2024 : 4389 - 4398 . doi: 10.1109/cvpr52733.2024.00420 http://dx.doi.org/10.1109/cvpr52733.2024.00420
TANG J , REN J , ZHOU H , et al . DreamGaussian: generative Gaussian splatting for efficient 3D content creation[DB/OL] . https://arxiv.org/abs/2309.16653 https://arxiv.org/abs/2309.16653 .
ERSHKOV V , PROSVIRYAKOV Y , BURMASHEVA V , et al . Towards understanding the algorithms for solving the Navier-Stokes equations [J]. Fluid Dynamics Research , 2021 , 53 ( 4 ): 044501 . doi: 10.1088/1873-7005/ac10f0 http://dx.doi.org/10.1088/1873-7005/ac10f0
YE T , PAN D , HUANG C , et al . Smoothed particle hydrodynamics (SPH) for complex fluid flows: recent developments in methodology and applications [J]. Physics of Fluids , 2019 , 31 ( 1 ): 011301 . doi: 10.1063/1.5068697 http://dx.doi.org/10.1063/1.5068697
THOMAS M , ALEX E , CHRISTOPH S , et al . Instant neural nraphics primitives with a multiresolution hash encoding [J]. ACM Transactions on Graphics , 2022 , 41 ( 4 ): 1 - 15 . doi: 10.1145/3528223.3530127 http://dx.doi.org/10.1145/3528223.3530127
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
鄂公网安备42010602004949号