SURFACE NORMAL RECONSTRUCTION USING POLARIZATION-UNET
Today, three-dimensional reconstruction of objects has many applications in various fields, and therefore, choosing a suitable method for high resolution three-dimensional reconstruction is an important issue and displaying high-level details in three-dimensional models is a serious challenge in thi...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2023-01-01
|
| Series: | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/X-4-W1-2022/537/2023/isprs-annals-X-4-W1-2022-537-2023.pdf |
| _version_ | 1797952275775225856 |
|---|---|
| author | F. S. Mortazavi S. Dajkhosh M. Saadatseresht |
| author_facet | F. S. Mortazavi S. Dajkhosh M. Saadatseresht |
| author_sort | F. S. Mortazavi |
| collection | DOAJ |
| description | Today, three-dimensional reconstruction of objects has many applications in various fields, and therefore, choosing a suitable method for high resolution three-dimensional reconstruction is an important issue and displaying high-level details in three-dimensional models is a serious challenge in this field. Until now, active methods have been used for high-resolution three-dimensional reconstruction. But the problem of active three-dimensional reconstruction methods is that they require a light source close to the object. Shape from polarization (SfP) is one of the best solutions for high-resolution three-dimensional reconstruction of objects, which is a passive method and does not have the drawbacks of active methods. The changes in polarization of the reflected light from an object can be analyzed by using a polarization camera or locating polarizing filter in front of the digital camera and rotating the filter. Using this information, the surface normal can be reconstructed with high accuracy, which will lead to local reconstruction of the surface details. In this paper, an end-to-end deep learning approach has been presented to produce the surface normal of objects. In this method a benchmark dataset has been used to train the neural network and evaluate the results. The results have been evaluated quantitatively and qualitatively by other methods and under different lighting conditions. The MAE value (Mean-Angular-Error) has been used for results evaluation. The evaluations showed that the proposed method could accurately reconstruct the surface normal of objects with the lowest MAE value which is equal to 18.06 degree on the whole dataset, in comparison to previous physics-based methods which are between 41.44 and 49.03 degree. |
| first_indexed | 2024-04-10T22:43:44Z |
| format | Article |
| id | doaj.art-47eb3e54f6c04b7b868d6b6b2343c50d |
| institution | Directory Open Access Journal |
| issn | 2194-9042 2194-9050 |
| language | English |
| last_indexed | 2024-04-10T22:43:44Z |
| publishDate | 2023-01-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj.art-47eb3e54f6c04b7b868d6b6b2343c50d2023-01-15T14:01:17ZengCopernicus PublicationsISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences2194-90422194-90502023-01-01X-4-W1-202253754310.5194/isprs-annals-X-4-W1-2022-537-2023SURFACE NORMAL RECONSTRUCTION USING POLARIZATION-UNETF. S. Mortazavi0S. Dajkhosh1M. Saadatseresht2School of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Tehran, IranSchool of Mathematics, Statistics and Computer Science, College of Science, University of Tehran, Tehran, IranSchool of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Tehran, IranToday, three-dimensional reconstruction of objects has many applications in various fields, and therefore, choosing a suitable method for high resolution three-dimensional reconstruction is an important issue and displaying high-level details in three-dimensional models is a serious challenge in this field. Until now, active methods have been used for high-resolution three-dimensional reconstruction. But the problem of active three-dimensional reconstruction methods is that they require a light source close to the object. Shape from polarization (SfP) is one of the best solutions for high-resolution three-dimensional reconstruction of objects, which is a passive method and does not have the drawbacks of active methods. The changes in polarization of the reflected light from an object can be analyzed by using a polarization camera or locating polarizing filter in front of the digital camera and rotating the filter. Using this information, the surface normal can be reconstructed with high accuracy, which will lead to local reconstruction of the surface details. In this paper, an end-to-end deep learning approach has been presented to produce the surface normal of objects. In this method a benchmark dataset has been used to train the neural network and evaluate the results. The results have been evaluated quantitatively and qualitatively by other methods and under different lighting conditions. The MAE value (Mean-Angular-Error) has been used for results evaluation. The evaluations showed that the proposed method could accurately reconstruct the surface normal of objects with the lowest MAE value which is equal to 18.06 degree on the whole dataset, in comparison to previous physics-based methods which are between 41.44 and 49.03 degree.https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/X-4-W1-2022/537/2023/isprs-annals-X-4-W1-2022-537-2023.pdf |
| spellingShingle | F. S. Mortazavi S. Dajkhosh M. Saadatseresht SURFACE NORMAL RECONSTRUCTION USING POLARIZATION-UNET ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | SURFACE NORMAL RECONSTRUCTION USING POLARIZATION-UNET |
| title_full | SURFACE NORMAL RECONSTRUCTION USING POLARIZATION-UNET |
| title_fullStr | SURFACE NORMAL RECONSTRUCTION USING POLARIZATION-UNET |
| title_full_unstemmed | SURFACE NORMAL RECONSTRUCTION USING POLARIZATION-UNET |
| title_short | SURFACE NORMAL RECONSTRUCTION USING POLARIZATION-UNET |
| title_sort | surface normal reconstruction using polarization unet |
| url | https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/X-4-W1-2022/537/2023/isprs-annals-X-4-W1-2022-537-2023.pdf |
| work_keys_str_mv | AT fsmortazavi surfacenormalreconstructionusingpolarizationunet AT sdajkhosh surfacenormalreconstructionusingpolarizationunet AT msaadatseresht surfacenormalreconstructionusingpolarizationunet |