132335 20260113073312.0 doi 10.1109/ICCV51070.2023.01945 sideral 137284 ART-2024-137284 eng Rodríguez-Puigvert, Javier Universidad de Zaragoza LightDepth: Single-view depth self-supervision from illumination decline 2024 Access copy available to the general public Unrestricted Single-view depth estimation can be remarkably effective if there is enough ground-truth depth data for supervised training. However, there are scenarios, especially in medicine in the case of endoscopies, where such data cannot be obtained. In such cases, multi-view self-supervision and synthetic-to-real transfer serve as alternative approaches, however, with a considerable performance reduction in comparison to supervised case. Instead, we propose a single-view self-supervised method that achieves a performance similar to the supervised case. In some medical devices, such as endoscopes, the camera and light sources are co-located at a small distance from the target surfaces. Thus, we can exploit that, for any given albedo and surface orientation, pixel brightness is inversely proportional to the square of the distance to the surface, providing a strong single-view self-supervisory signal. In our experiments, our self-supervised models deliver accuracies comparable to those of fully supervised ones, while being applicable without depth ground-truth data. info:eu-repo/grantAgreement/ES/DGA/T45-23R info:eu-repo/grantAgreement/EC/H2020/863146/EU/EndoMapper: Real-time mapping from endoscopic video/EndoMapper This project has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No H2020 863146-EndoMapper info:eu-repo/grantAgreement/ES/MCIU-AEI-FEDER/PGC2018-096367-B-I00 info:eu-repo/grantAgreement/ES/MCIU/FPU20-0678 info:eu-repo/grantAgreement/ES/MICINN-AEI/PID2021-125209OB-I00 info:eu-repo/grantAgreement/ES/MICINN/PID2021-127685NB-I00 info:eu-repo/grantAgreement/EUR/MICINN/TED2021-131150B-I00 info:eu-repo/semantics/openAccess All rights reserved http://www.europeana.eu/rights/rr-f/ 3.544 2024 Software 2024 Computer Vision and Pattern Recognition 2024 38.1 2024 info:eu-repo/semantics/conferenceObject info:eu-repo/semantics/acceptedVersion (orcid)0000-0002-6837-934X Batlle, Víctor M. Universidad de Zaragoza (orcid)0000-0002-3627-7306 Martínez Montiel, J.M. Universidad de Zaragoza (orcid)0000-0002-6741-844X Martinez-Cantin, Ruben Universidad de Zaragoza Fua, Pascal (orcid)0000-0002-4518-5876 Tardós, Juan D. Universidad de Zaragoza (orcid)0000-0003-1368-1151 Civera, Javier Universidad de Zaragoza 5007 520 Universidad de Zaragoza Dpto. Informát.Ingenie.Sistms. Área Ingen.Sistemas y Automát. 2023 (2024), 21216-21226 Proceedings (IEEE Int. Conf. Comput. Vision) Proceedings (IEEE International Conference on Computer Vision) 1550-5499 2029649 http://zaguan.unizar.es/record/132335/files/texto_completo.pdf Postprint info:eu-repo/date/embargoEnd/2025-01-15 2509136 http://zaguan.unizar.es/record/132335/files/texto_completo.jpg?subformat=icon icon Postprint info:eu-repo/date/embargoEnd/2025-01-15 oai:zaguan.unizar.es:132335 articulos driver 2026-01-12-12:48:00 ARTICLE