Uncertainty Quantification of Neural Reflectance Fields for Underwater Scenes
Neural radiance fields and neural reflectance fields are novel deep learning methods for generating novel views of 3D scenes from 2D images. To extend the neural scene representation techniques to complex underwater environments, beyond neural reflectance fields underwater (BNU) was proposed, which...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Journal of Marine Science and Engineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-1312/12/2/349 |
| _version_ | 1797297737493905408 |
|---|---|
| author | Haojie Lian Xinhao Li Leilei Chen Xin Wen Mengxi Zhang Jieyuan Zhang Yilin Qu |
| author_facet | Haojie Lian Xinhao Li Leilei Chen Xin Wen Mengxi Zhang Jieyuan Zhang Yilin Qu |
| author_sort | Haojie Lian |
| collection | DOAJ |
| description | Neural radiance fields and neural reflectance fields are novel deep learning methods for generating novel views of 3D scenes from 2D images. To extend the neural scene representation techniques to complex underwater environments, beyond neural reflectance fields underwater (BNU) was proposed, which considers the relighting conditions of on-aboard light sources by using neural reflectance fields, and approximates the attenuation and backscatter effects of water with an additional constant. Because the quality of the neural representation of underwater scenes is critical to downstream tasks such as marine surveying and mapping, the model reliability should be considered and evaluated. However, current neural reflectance models lack the ability of quantifying the uncertainty of underwater scenes that are not directly observed during training, which hinders their widespread use in the field of underwater unmanned autonomous navigation. To address this issue, we introduce an ensemble strategy to BNU that quantifies cognitive uncertainty in color space and unobserved regions with the expectation and variance of RGB values and termination probabilities along the ray. We also employ a regularization method to smooth the density of the underwater neural reflectance model. The effectiveness of the present method is demonstrated in numerical experiments. |
| first_indexed | 2024-03-07T22:25:39Z |
| format | Article |
| id | doaj.art-601c0be516874326867766605ad7d4e5 |
| institution | Directory Open Access Journal |
| issn | 2077-1312 |
| language | English |
| last_indexed | 2024-03-07T22:25:39Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Marine Science and Engineering |
| spelling | doaj.art-601c0be516874326867766605ad7d4e52024-02-23T15:23:22ZengMDPI AGJournal of Marine Science and Engineering2077-13122024-02-0112234910.3390/jmse12020349Uncertainty Quantification of Neural Reflectance Fields for Underwater ScenesHaojie Lian0Xinhao Li1Leilei Chen2Xin Wen3Mengxi Zhang4Jieyuan Zhang5Yilin Qu6Key Laboratory of In-Situ Property-Improving Mining of Ministry of Education, Taiyuan University of Technology, Taiyuan 030024, ChinaKey Laboratory of In-Situ Property-Improving Mining of Ministry of Education, Taiyuan University of Technology, Taiyuan 030024, ChinaHenan International Joint Laboratory of Structural Mechanics and Computational Simulation, School of Architecture and Civil Engineering, Huanghuai University, Zhumadian 463000, ChinaSchool of Software, Taiyuan University of Technology, Jinzhong 030600, ChinaState Key Laboratory of Hydraulic Engineering Intelligent Construction and Operation, Tianjin University, Tianjin 300350, ChinaAcademy of Military Science, Beijing 100091, ChinaSchool of Marine Science and Technology, Northwestern Polytechnical University, Xi’an 710072, ChinaNeural radiance fields and neural reflectance fields are novel deep learning methods for generating novel views of 3D scenes from 2D images. To extend the neural scene representation techniques to complex underwater environments, beyond neural reflectance fields underwater (BNU) was proposed, which considers the relighting conditions of on-aboard light sources by using neural reflectance fields, and approximates the attenuation and backscatter effects of water with an additional constant. Because the quality of the neural representation of underwater scenes is critical to downstream tasks such as marine surveying and mapping, the model reliability should be considered and evaluated. However, current neural reflectance models lack the ability of quantifying the uncertainty of underwater scenes that are not directly observed during training, which hinders their widespread use in the field of underwater unmanned autonomous navigation. To address this issue, we introduce an ensemble strategy to BNU that quantifies cognitive uncertainty in color space and unobserved regions with the expectation and variance of RGB values and termination probabilities along the ray. We also employ a regularization method to smooth the density of the underwater neural reflectance model. The effectiveness of the present method is demonstrated in numerical experiments.https://www.mdpi.com/2077-1312/12/2/349neural reflectance fieldsunderwater scenesuncertainty quantification |
| spellingShingle | Haojie Lian Xinhao Li Leilei Chen Xin Wen Mengxi Zhang Jieyuan Zhang Yilin Qu Uncertainty Quantification of Neural Reflectance Fields for Underwater Scenes Journal of Marine Science and Engineering neural reflectance fields underwater scenes uncertainty quantification |
| title | Uncertainty Quantification of Neural Reflectance Fields for Underwater Scenes |
| title_full | Uncertainty Quantification of Neural Reflectance Fields for Underwater Scenes |
| title_fullStr | Uncertainty Quantification of Neural Reflectance Fields for Underwater Scenes |
| title_full_unstemmed | Uncertainty Quantification of Neural Reflectance Fields for Underwater Scenes |
| title_short | Uncertainty Quantification of Neural Reflectance Fields for Underwater Scenes |
| title_sort | uncertainty quantification of neural reflectance fields for underwater scenes |
| topic | neural reflectance fields underwater scenes uncertainty quantification |
| url | https://www.mdpi.com/2077-1312/12/2/349 |
| work_keys_str_mv | AT haojielian uncertaintyquantificationofneuralreflectancefieldsforunderwaterscenes AT xinhaoli uncertaintyquantificationofneuralreflectancefieldsforunderwaterscenes AT leileichen uncertaintyquantificationofneuralreflectancefieldsforunderwaterscenes AT xinwen uncertaintyquantificationofneuralreflectancefieldsforunderwaterscenes AT mengxizhang uncertaintyquantificationofneuralreflectancefieldsforunderwaterscenes AT jieyuanzhang uncertaintyquantificationofneuralreflectancefieldsforunderwaterscenes AT yilinqu uncertaintyquantificationofneuralreflectancefieldsforunderwaterscenes |