TanDepth: Leveraging Global DEMs for Metric Monocular Depth Estimation in UAVs
Aerial scene understanding systems face stringent payload restrictions and must often rely on monocular depth estimation for modeling scene geometry, which is an inherently ill-posed problem. Moreover, obtaining accurate ground truth data required by learning-based methods raises significant additio...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10848130/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850023624543567872 |
|---|---|
| author | Horatiu Florea Sergiu Nedevschi |
| author_facet | Horatiu Florea Sergiu Nedevschi |
| author_sort | Horatiu Florea |
| collection | DOAJ |
| description | Aerial scene understanding systems face stringent payload restrictions and must often rely on monocular depth estimation for modeling scene geometry, which is an inherently ill-posed problem. Moreover, obtaining accurate ground truth data required by learning-based methods raises significant additional challenges in the aerial domain. Self-supervised approaches can bypass this problem, at the cost of providing only up-to-scale results. Similarly, recent supervised solutions which make good progress toward zero-shot generalization also provide only relative depth values. This work presents TanDepth, a practical scale recovery method for obtaining metric depth results from relative estimations at inference-time, irrespective of the type of model generating them. Tailored for uncrewed aerial vehicle (UAV) applications, our method leverages sparse measurements from Global Digital Elevation Models (GDEM) by projecting them to the camera view using extrinsic and intrinsic information. An adaptation to the cloth simulation filter is presented, which allows selecting ground points from the estimated depth map to then correlate with the projected reference points. We evaluate and compare our method against alternate scaling methods adapted for UAVs, on a variety of real-world scenes. Considering the limited availability of data for this domain, we construct and release a comprehensive, depth-focused extension to the popular UAVid dataset to further research. |
| format | Article |
| id | doaj-art-3977cb7ecd904473b9aebaef61d9be20 |
| institution | DOAJ |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-3977cb7ecd904473b9aebaef61d9be202025-08-20T03:01:19ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01185445545910.1109/JSTARS.2025.353198410848130TanDepth: Leveraging Global DEMs for Metric Monocular Depth Estimation in UAVsHoratiu Florea0https://orcid.org/0000-0003-0502-360XSergiu Nedevschi1https://orcid.org/0000-0003-2018-4647Department of Computer Science, Technical University of Cluj-Napoca, Cluj-Napoca, RomaniaDepartment of Computer Science, Technical University of Cluj-Napoca, Cluj-Napoca, RomaniaAerial scene understanding systems face stringent payload restrictions and must often rely on monocular depth estimation for modeling scene geometry, which is an inherently ill-posed problem. Moreover, obtaining accurate ground truth data required by learning-based methods raises significant additional challenges in the aerial domain. Self-supervised approaches can bypass this problem, at the cost of providing only up-to-scale results. Similarly, recent supervised solutions which make good progress toward zero-shot generalization also provide only relative depth values. This work presents TanDepth, a practical scale recovery method for obtaining metric depth results from relative estimations at inference-time, irrespective of the type of model generating them. Tailored for uncrewed aerial vehicle (UAV) applications, our method leverages sparse measurements from Global Digital Elevation Models (GDEM) by projecting them to the camera view using extrinsic and intrinsic information. An adaptation to the cloth simulation filter is presented, which allows selecting ground points from the estimated depth map to then correlate with the projected reference points. We evaluate and compare our method against alternate scaling methods adapted for UAVs, on a variety of real-world scenes. Considering the limited availability of data for this domain, we construct and release a comprehensive, depth-focused extension to the popular UAVid dataset to further research.https://ieeexplore.ieee.org/document/10848130/Aerial scene understandingglobal digital elevation models (GDEMs)metric monocular depth estimation (MDE) |
| spellingShingle | Horatiu Florea Sergiu Nedevschi TanDepth: Leveraging Global DEMs for Metric Monocular Depth Estimation in UAVs IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Aerial scene understanding global digital elevation models (GDEMs) metric monocular depth estimation (MDE) |
| title | TanDepth: Leveraging Global DEMs for Metric Monocular Depth Estimation in UAVs |
| title_full | TanDepth: Leveraging Global DEMs for Metric Monocular Depth Estimation in UAVs |
| title_fullStr | TanDepth: Leveraging Global DEMs for Metric Monocular Depth Estimation in UAVs |
| title_full_unstemmed | TanDepth: Leveraging Global DEMs for Metric Monocular Depth Estimation in UAVs |
| title_short | TanDepth: Leveraging Global DEMs for Metric Monocular Depth Estimation in UAVs |
| title_sort | tandepth leveraging global dems for metric monocular depth estimation in uavs |
| topic | Aerial scene understanding global digital elevation models (GDEMs) metric monocular depth estimation (MDE) |
| url | https://ieeexplore.ieee.org/document/10848130/ |
| work_keys_str_mv | AT horatiuflorea tandepthleveragingglobaldemsformetricmonoculardepthestimationinuavs AT sergiunedevschi tandepthleveragingglobaldemsformetricmonoculardepthestimationinuavs |