In-the-Wild Single Camera 3D Reconstruction Through Moving Water Surfaces
Jinhui Xiong, Wolfgang Heidrich
Abstract
We present a method for reconstructing the 3D shape of underwater environments from a single, stationary camera placed above the water. We propose a novel differentiable framework, which, to our knowledge, is the first single-camera solution that is capable of simultaneously retrieving the structure of dynamic water surfaces and static underwater scene geometry in the wild. This framework integrates ray casting of Snell’s law at the refractive interface, multi-view triangulation and specially designed loss functions.Our method is calibration-free, and thus it is easy to collect data outdoors in uncontrolled environments. Experimental results show that our method is able to realize robust and quality reconstructions on a variety of scenes, both in a laboratory environment and in the wild, and even in a salt water environment. We believe the method is promising for applications in surveying and environmental monitoring.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext 5e21de38-a15a-4608-8d70-36918cdaa013Cited by top-tier papers4
- NeRFrac: Neural Radiance Fields through Refractive SurfaceYifan Zhan, Shohei Nobuhara, Ko Nishino, Yinqiang ZhengICCV 2023 · 18 citations
- Gravitationally Lensed Black Hole Emission TomographyAviad Levis, Pratul P. Srinivasan, Andrew A. Chael, Ren Ng et al.CVPR 2022 · 10 citations
- Looking Into the Water by Unsupervised Learning of the Surface ShapeOri Lifschitz, Tali Treibitz, Dan RosenbaumNeurIPS 2025 · 1 citation
- Dynamic Black-hole Emission Tomography with Physics-informed Neural FieldsBerthy T. Feng, Andrew A. Chael, David Bromley, Aviad Levis et al.CVPR 2026 · 1 citation
Builds on3
- Restoration of Non-Rigidly Distorted Underwater Images Using a Combination of Compressive Sensing and Local Polynomial Image RepresentationsJerin Geo James, Pranay Agrawal, Ajit RajwadeICCV 2019 · 23 citations
- Through the Looking Glass: Neural 3D Reconstruction of Transparent ShapesZhengqin Li, Yu-Ying Yeh, Manmohan ChandrakerCVPR 2020
- Dynamic Fluid Surface Reconstruction Using Deep Neural NetworkSimron Thapa, Nianyi Li, Jinwei YeCVPR 2020
Related papers
- Surface Normals and Shape From WaterSatoshi Murai, Meng-Yu Kuo, Ryo Kawahara, Shohei Nobuhara et al.ICCV 2019 · 14 citations
- Neural Underwater Scene RepresentationYunkai Tang, Chengxuan Zhu, Renjie Wan, Chao Xu et al.CVPR 2024 · 19 citations
- Differentiable Neural Surface Refinement for Modeling Transparent ObjectsWeijian Deng, Dylan Campbell, Chunyi Sun, Shubham Kanitkar et al.CVPR 2024
- Neural Surface Reconstruction of Dynamic Scenes with Monocular RGB-D CameraHongrui Cai, Wanquan Feng, Xuetao Feng, Yan Wang et al.NeurIPS 2022 · 83 citations
- DiffTrans: Differentiable Geometry-Materials Decomposition for Reconstructing Transparent ObjectsChangpu Li, Shuang Wu, Songlin Tang, Guangming Lu et al.ICLR 2026
