Explaining Monocular Depth Estimation - Diving into Regional Differences in the Prediction
Norsk IKT-konferanse for forskning og utdanning · 2025
DOI: 10.5324/3kmsk121 ↗Kurz gefasst
Fachartikel in der Zeitschrift Norsk IKT-konferanse for forskning og utdanning, 2025. Verfasst von Sabine Fischer und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Recent foundational depth estimation models achieve impressive accuracy on various scenes. However, due to their black box nature, we lack knowledge about how they utilize the input for their predictions, and hence about their applicability to domains sparsely covered by labeled datasets. This paper, therefore, applies occlusion to investigate the influence of the input on different parts of the predicted depth maps for underwater scenes. Our results show that foundational depth estimation models combine global and local features to estimate relative distance, and that their influence differs significantly between the background and the rest of the scene. This suggests that extending post-hoc explanations to consider relationships between multiple input and output features can enrich our understanding of monocular depth estimation models and potentially help to gauge their applicability to new domains.
Zitieren
APA
Fischer, S., Ødegård, Ø., Lindseth, F., Moan, S. Y. L., & Kiss, G. H. (2025). Explaining Monocular Depth Estimation - Diving into Regional Differences in the Prediction. Norsk IKT-konferanse for forskning og utdanning. https://doi.org/10.5324/3kmsk121
Vancouver
Fischer S, Ødegård Ø, Lindseth F, Moan SYL, Kiss GH. Explaining Monocular Depth Estimation - Diving into Regional Differences in the Prediction. Norsk IKT-konferanse for forskning og utdanning. 2025. doi:10.5324/3kmsk121
BibTeX
@article{Fischer2025,
author = {Fischer, Sabine and Ødegård, Øyvind and Lindseth, Frank and Moan, Steven Yves Le and Kiss, Gabriel Hanssen},
title = {Explaining Monocular Depth Estimation - Diving into Regional Differences in the Prediction},
journal = {Norsk IKT-konferanse for forskning og utdanning},
year = {2025},
doi = {10.5324/3kmsk121}
}
EUVIMED