Triplane generator-based NeRF-GAN framework for single-view ship reconstruction

Acquiring sufficient visual information for the three-dimensional (3D) reconstruction of ships in navigation is particularly challenging. With the evolution of 3D reconstruction methodologies predicated on neural rendering, the computational pipeline for 3D reconstruction has undergone enhancements...

Full description

Saved in:
Bibliographic Details
Main Authors: Tao Liu, Shiqi Geng, Yucheng Fu, Zhengling Lei, Yuchi Huo, Xiaocai Zhang, Fang Wang, Bing Han, Mei Sha, Zhongdai Wu
Format: Article
Language:English
Published: Taylor & Francis Group 2025-08-01
Series:International Journal of Digital Earth
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/17538947.2025.2496406
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849224263379189760
author Tao Liu
Shiqi Geng
Yucheng Fu
Zhengling Lei
Yuchi Huo
Xiaocai Zhang
Fang Wang
Bing Han
Mei Sha
Zhongdai Wu
author_facet Tao Liu
Shiqi Geng
Yucheng Fu
Zhengling Lei
Yuchi Huo
Xiaocai Zhang
Fang Wang
Bing Han
Mei Sha
Zhongdai Wu
author_sort Tao Liu
collection DOAJ
description Acquiring sufficient visual information for the three-dimensional (3D) reconstruction of ships in navigation is particularly challenging. With the evolution of 3D reconstruction methodologies predicated on neural rendering, the computational pipeline for 3D reconstruction has undergone enhancements and optimizations. However, this pipeline necessitates a substantial corpus of input images. Research into 3D reconstruction from monocular images is in its nascent stages, and to date, no unsupervised deep learning approach for 3D reconstruction of ships from single-view UAV imagery exists within the realm of navigation. This paper introduces a novel network architecture for reconstructing 3D representations of ships from single-view UAV images. Initially, a priori statistical analysis of the dataset is conducted to harness color distribution information for noise generation. Subsequently, a novel generator and mask module are engineered to produce optimized feature outputs. Plus, discriminator and encoder networks, coupled with a tailored loss function, are formulated to direct model optimization. Ultimately, to demonstrate the effectiveness of our proposed method for single-view 3D reconstruction, we conducted experiments across three distinct datasets from various domains. Our method's FID value of 10.61 is impressive. At the same time, it yields an LPIPS value of 0.091, which is the best among the six different methods.
format Article
id doaj-art-01a6f9eab800419182a7951631653ff9
institution Kabale University
issn 1753-8947
1753-8955
language English
publishDate 2025-08-01
publisher Taylor & Francis Group
record_format Article
series International Journal of Digital Earth
spelling doaj-art-01a6f9eab800419182a7951631653ff92025-08-25T11:28:16ZengTaylor & Francis GroupInternational Journal of Digital Earth1753-89471753-89552025-08-0118110.1080/17538947.2025.2496406Triplane generator-based NeRF-GAN framework for single-view ship reconstructionTao Liu0Shiqi Geng1Yucheng Fu2Zhengling Lei3Yuchi Huo4Xiaocai Zhang5Fang Wang6Bing Han7Mei Sha8Zhongdai Wu9College of Transport & Communications, Shanghai Maritime University, Shanghai, People’s Republic of ChinaCollege of Transport & Communications, Shanghai Maritime University, Shanghai, People’s Republic of ChinaCollege of Transport & Communications, Shanghai Maritime University, Shanghai, People’s Republic of ChinaCollege of Engineering Science and Technology, Shanghai Ocean University, Shanghai, People’s Republic of ChinaState Key Lab of CAD&CG, Zhejiang University, Hangzhou, People’s Republic of ChinaFaculty of Engineering and Information Technology, University of Melbourne, Parkville, AustraliaCollege of Engineering Science and Technology, Shanghai Ocean University, Shanghai, People’s Republic of ChinaState Key Laboratory of Maritime Technology and Safety, Shanghai Ship and Shipping Research Institute Co., Ltd., Shanghai, People’s Republic of ChinaCollege of Transport & Communications, Shanghai Maritime University, Shanghai, People’s Republic of ChinaState Key Laboratory of Maritime Technology and Safety, Shanghai Ship and Shipping Research Institute Co., Ltd., Shanghai, People’s Republic of ChinaAcquiring sufficient visual information for the three-dimensional (3D) reconstruction of ships in navigation is particularly challenging. With the evolution of 3D reconstruction methodologies predicated on neural rendering, the computational pipeline for 3D reconstruction has undergone enhancements and optimizations. However, this pipeline necessitates a substantial corpus of input images. Research into 3D reconstruction from monocular images is in its nascent stages, and to date, no unsupervised deep learning approach for 3D reconstruction of ships from single-view UAV imagery exists within the realm of navigation. This paper introduces a novel network architecture for reconstructing 3D representations of ships from single-view UAV images. Initially, a priori statistical analysis of the dataset is conducted to harness color distribution information for noise generation. Subsequently, a novel generator and mask module are engineered to produce optimized feature outputs. Plus, discriminator and encoder networks, coupled with a tailored loss function, are formulated to direct model optimization. Ultimately, to demonstrate the effectiveness of our proposed method for single-view 3D reconstruction, we conducted experiments across three distinct datasets from various domains. Our method's FID value of 10.61 is impressive. At the same time, it yields an LPIPS value of 0.091, which is the best among the six different methods.https://www.tandfonline.com/doi/10.1080/17538947.2025.24964063D Environment perceptionsingle view ship reconstructionmaritime digital twinneural radiance fieldgenerative adversarial network
spellingShingle Tao Liu
Shiqi Geng
Yucheng Fu
Zhengling Lei
Yuchi Huo
Xiaocai Zhang
Fang Wang
Bing Han
Mei Sha
Zhongdai Wu
Triplane generator-based NeRF-GAN framework for single-view ship reconstruction
International Journal of Digital Earth
3D Environment perception
single view ship reconstruction
maritime digital twin
neural radiance field
generative adversarial network
title Triplane generator-based NeRF-GAN framework for single-view ship reconstruction
title_full Triplane generator-based NeRF-GAN framework for single-view ship reconstruction
title_fullStr Triplane generator-based NeRF-GAN framework for single-view ship reconstruction
title_full_unstemmed Triplane generator-based NeRF-GAN framework for single-view ship reconstruction
title_short Triplane generator-based NeRF-GAN framework for single-view ship reconstruction
title_sort triplane generator based nerf gan framework for single view ship reconstruction
topic 3D Environment perception
single view ship reconstruction
maritime digital twin
neural radiance field
generative adversarial network
url https://www.tandfonline.com/doi/10.1080/17538947.2025.2496406
work_keys_str_mv AT taoliu triplanegeneratorbasednerfganframeworkforsingleviewshipreconstruction
AT shiqigeng triplanegeneratorbasednerfganframeworkforsingleviewshipreconstruction
AT yuchengfu triplanegeneratorbasednerfganframeworkforsingleviewshipreconstruction
AT zhenglinglei triplanegeneratorbasednerfganframeworkforsingleviewshipreconstruction
AT yuchihuo triplanegeneratorbasednerfganframeworkforsingleviewshipreconstruction
AT xiaocaizhang triplanegeneratorbasednerfganframeworkforsingleviewshipreconstruction
AT fangwang triplanegeneratorbasednerfganframeworkforsingleviewshipreconstruction
AT binghan triplanegeneratorbasednerfganframeworkforsingleviewshipreconstruction
AT meisha triplanegeneratorbasednerfganframeworkforsingleviewshipreconstruction
AT zhongdaiwu triplanegeneratorbasednerfganframeworkforsingleviewshipreconstruction