A Critical Analysis of NeRF-Based 3D Reconstruction
This paper presents a critical analysis of image-based 3D reconstruction using neural radiance fields (NeRFs), with a focus on quantitative comparisons with respect to traditional photogrammetry. The aim is, therefore, to objectively evaluate the strengths and weaknesses of NeRFs and provide insight...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-07-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/14/3585 |
| _version_ | 1797587600349855744 |
|---|---|
| author | Fabio Remondino Ali Karami Ziyang Yan Gabriele Mazzacca Simone Rigon Rongjun Qin |
| author_facet | Fabio Remondino Ali Karami Ziyang Yan Gabriele Mazzacca Simone Rigon Rongjun Qin |
| author_sort | Fabio Remondino |
| collection | DOAJ |
| description | This paper presents a critical analysis of image-based 3D reconstruction using neural radiance fields (NeRFs), with a focus on quantitative comparisons with respect to traditional photogrammetry. The aim is, therefore, to objectively evaluate the strengths and weaknesses of NeRFs and provide insights into their applicability to different real-life scenarios, from small objects to heritage and industrial scenes. After a comprehensive overview of photogrammetry and NeRF methods, highlighting their respective advantages and disadvantages, various NeRF methods are compared using diverse objects with varying sizes and surface characteristics, including texture-less, metallic, translucent, and transparent surfaces. We evaluated the quality of the resulting 3D reconstructions using multiple criteria, such as noise level, geometric accuracy, and the number of required images (i.e., image baselines). The results show that NeRFs exhibit superior performance over photogrammetry in terms of non-collaborative objects with texture-less, reflective, and refractive surfaces. Conversely, photogrammetry outperforms NeRFs in cases where the object’s surface possesses cooperative texture. Such complementarity should be further exploited in future works. |
| first_indexed | 2024-03-11T00:41:13Z |
| format | Article |
| id | doaj.art-9d8a453d24ce4b278b83df05cc948520 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T00:41:13Z |
| publishDate | 2023-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-9d8a453d24ce4b278b83df05cc9485202023-11-18T21:12:52ZengMDPI AGRemote Sensing2072-42922023-07-011514358510.3390/rs15143585A Critical Analysis of NeRF-Based 3D ReconstructionFabio Remondino0Ali Karami1Ziyang Yan2Gabriele Mazzacca3Simone Rigon4Rongjun Qin53D Optical Metrology Unit, Bruno Kessler Foundation (FBK), Via Sommarive 18, 38123 Trento, Italy3D Optical Metrology Unit, Bruno Kessler Foundation (FBK), Via Sommarive 18, 38123 Trento, Italy3D Optical Metrology Unit, Bruno Kessler Foundation (FBK), Via Sommarive 18, 38123 Trento, Italy3D Optical Metrology Unit, Bruno Kessler Foundation (FBK), Via Sommarive 18, 38123 Trento, Italy3D Optical Metrology Unit, Bruno Kessler Foundation (FBK), Via Sommarive 18, 38123 Trento, ItalyGeospatial Data Analytics Laboratory, The Ohio State University, 218B Bolz Hall, 2036 Neil Avenue, Columbus, OH 43210, USAThis paper presents a critical analysis of image-based 3D reconstruction using neural radiance fields (NeRFs), with a focus on quantitative comparisons with respect to traditional photogrammetry. The aim is, therefore, to objectively evaluate the strengths and weaknesses of NeRFs and provide insights into their applicability to different real-life scenarios, from small objects to heritage and industrial scenes. After a comprehensive overview of photogrammetry and NeRF methods, highlighting their respective advantages and disadvantages, various NeRF methods are compared using diverse objects with varying sizes and surface characteristics, including texture-less, metallic, translucent, and transparent surfaces. We evaluated the quality of the resulting 3D reconstructions using multiple criteria, such as noise level, geometric accuracy, and the number of required images (i.e., image baselines). The results show that NeRFs exhibit superior performance over photogrammetry in terms of non-collaborative objects with texture-less, reflective, and refractive surfaces. Conversely, photogrammetry outperforms NeRFs in cases where the object’s surface possesses cooperative texture. Such complementarity should be further exploited in future works.https://www.mdpi.com/2072-4292/15/14/3585photogrammetryneural radiance fieldsNeRF3D reconstructionqualityaccuracy |
| spellingShingle | Fabio Remondino Ali Karami Ziyang Yan Gabriele Mazzacca Simone Rigon Rongjun Qin A Critical Analysis of NeRF-Based 3D Reconstruction Remote Sensing photogrammetry neural radiance fields NeRF 3D reconstruction quality accuracy |
| title | A Critical Analysis of NeRF-Based 3D Reconstruction |
| title_full | A Critical Analysis of NeRF-Based 3D Reconstruction |
| title_fullStr | A Critical Analysis of NeRF-Based 3D Reconstruction |
| title_full_unstemmed | A Critical Analysis of NeRF-Based 3D Reconstruction |
| title_short | A Critical Analysis of NeRF-Based 3D Reconstruction |
| title_sort | critical analysis of nerf based 3d reconstruction |
| topic | photogrammetry neural radiance fields NeRF 3D reconstruction quality accuracy |
| url | https://www.mdpi.com/2072-4292/15/14/3585 |
| work_keys_str_mv | AT fabioremondino acriticalanalysisofnerfbased3dreconstruction AT alikarami acriticalanalysisofnerfbased3dreconstruction AT ziyangyan acriticalanalysisofnerfbased3dreconstruction AT gabrielemazzacca acriticalanalysisofnerfbased3dreconstruction AT simonerigon acriticalanalysisofnerfbased3dreconstruction AT rongjunqin acriticalanalysisofnerfbased3dreconstruction AT fabioremondino criticalanalysisofnerfbased3dreconstruction AT alikarami criticalanalysisofnerfbased3dreconstruction AT ziyangyan criticalanalysisofnerfbased3dreconstruction AT gabrielemazzacca criticalanalysisofnerfbased3dreconstruction AT simonerigon criticalanalysisofnerfbased3dreconstruction AT rongjunqin criticalanalysisofnerfbased3dreconstruction |