MONOCULAR DEPTH PREDICTION IN PHOTOGRAMMETRIC APPLICATIONS
Despite the recent success of learning-based monocular depth estimation algorithms and the release of large-scale datasets for training, the methods are limited to depth map prediction and still struggle to yield reliable results in the 3D space without additional scene cues. Indeed, although state-...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2022-05-01
|
| Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLIII-B2-2022/469/2022/isprs-archives-XLIII-B2-2022-469-2022.pdf |
| _version_ | 1828346455948001280 |
|---|---|
| author | M. Welponer E. K. Stathopoulou F. Remondino |
| author_facet | M. Welponer E. K. Stathopoulou F. Remondino |
| author_sort | M. Welponer |
| collection | DOAJ |
| description | Despite the recent success of learning-based monocular depth estimation algorithms and the release of large-scale datasets for training, the methods are limited to depth map prediction and still struggle to yield reliable results in the 3D space without additional scene cues. Indeed, although state-of-the-art approaches produce quality depth maps, they generally fail to recover the 3D structure of the scene robustly. This work explores supervised CNN architectures for monocular depth estimation and evaluates their potential in 3D reconstruction. Since most available datasets for training are not designed toward this goal and are limited to specific indoor scenarios, a new metric, large-scale synthetic benchmark (ArchDepth) is introduced that renders near real-world scenarios of outdoor scenes. A encoder-decoder architecture is used for training, and the generalization of the approach is evaluated via depth inference in unseen views in synthetic and real-world scenarios. The depth map predictions are also projected in the 3D space using a separate module. Results are qualitatively and quantitatively evaluated and compared with state-of-the-art algorithms for single image 3D scene recovery. |
| first_indexed | 2024-04-14T00:26:40Z |
| format | Article |
| id | doaj.art-b2314d86ce5a46529b666d6056b666c8 |
| institution | Directory Open Access Journal |
| issn | 1682-1750 2194-9034 |
| language | English |
| last_indexed | 2024-04-14T00:26:40Z |
| publishDate | 2022-05-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj.art-b2314d86ce5a46529b666d6056b666c82022-12-22T02:22:41ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342022-05-01XLIII-B2-202246947610.5194/isprs-archives-XLIII-B2-2022-469-2022MONOCULAR DEPTH PREDICTION IN PHOTOGRAMMETRIC APPLICATIONSM. Welponer0E. K. Stathopoulou1F. Remondino23D Optical Metrology (3DOM) Unit, Bruno Kessler Foundation (FBK), Trento, Italy3D Optical Metrology (3DOM) Unit, Bruno Kessler Foundation (FBK), Trento, Italy3D Optical Metrology (3DOM) Unit, Bruno Kessler Foundation (FBK), Trento, ItalyDespite the recent success of learning-based monocular depth estimation algorithms and the release of large-scale datasets for training, the methods are limited to depth map prediction and still struggle to yield reliable results in the 3D space without additional scene cues. Indeed, although state-of-the-art approaches produce quality depth maps, they generally fail to recover the 3D structure of the scene robustly. This work explores supervised CNN architectures for monocular depth estimation and evaluates their potential in 3D reconstruction. Since most available datasets for training are not designed toward this goal and are limited to specific indoor scenarios, a new metric, large-scale synthetic benchmark (ArchDepth) is introduced that renders near real-world scenarios of outdoor scenes. A encoder-decoder architecture is used for training, and the generalization of the approach is evaluated via depth inference in unseen views in synthetic and real-world scenarios. The depth map predictions are also projected in the 3D space using a separate module. Results are qualitatively and quantitatively evaluated and compared with state-of-the-art algorithms for single image 3D scene recovery.https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLIII-B2-2022/469/2022/isprs-archives-XLIII-B2-2022-469-2022.pdf |
| spellingShingle | M. Welponer E. K. Stathopoulou F. Remondino MONOCULAR DEPTH PREDICTION IN PHOTOGRAMMETRIC APPLICATIONS The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | MONOCULAR DEPTH PREDICTION IN PHOTOGRAMMETRIC APPLICATIONS |
| title_full | MONOCULAR DEPTH PREDICTION IN PHOTOGRAMMETRIC APPLICATIONS |
| title_fullStr | MONOCULAR DEPTH PREDICTION IN PHOTOGRAMMETRIC APPLICATIONS |
| title_full_unstemmed | MONOCULAR DEPTH PREDICTION IN PHOTOGRAMMETRIC APPLICATIONS |
| title_short | MONOCULAR DEPTH PREDICTION IN PHOTOGRAMMETRIC APPLICATIONS |
| title_sort | monocular depth prediction in photogrammetric applications |
| url | https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLIII-B2-2022/469/2022/isprs-archives-XLIII-B2-2022-469-2022.pdf |
| work_keys_str_mv | AT mwelponer monoculardepthpredictioninphotogrammetricapplications AT ekstathopoulou monoculardepthpredictioninphotogrammetricapplications AT fremondino monoculardepthpredictioninphotogrammetricapplications |