001034963 001__ 1034963
001034963 005__ 20250203103113.0
001034963 037__ $$aFZJ-2025-00071
001034963 041__ $$aEnglish
001034963 1001_ $$0P:(DE-Juel1)188471$$aQuercia, Alessio$$b0$$eCorresponding author$$ufzj
001034963 1112_ $$aHelmholtz AI Conference$$cDüsseldorf$$d2024-06-12 - 2024-06-14$$wGermany
001034963 245__ $$aMulti-Source Auxiliary Tasks supported Monocular Depth Estimation
001034963 260__ $$c2024
001034963 3367_ $$033$$2EndNote$$aConference Paper
001034963 3367_ $$2DataCite$$aOther
001034963 3367_ $$2BibTeX$$aINPROCEEDINGS
001034963 3367_ $$2DRIVER$$aconferenceObject
001034963 3367_ $$2ORCID$$aLECTURE_SPEECH
001034963 3367_ $$0PUB:(DE-HGF)6$$2PUB:(DE-HGF)$$aConference Presentation$$bconf$$mconf$$s1736773569_16364$$xAfter Call
001034963 500__ $$aThe original abstract contains figures that cannot be shown here.
001034963 520__ $$aMonocular depth estimation (MDE) is a challenging task in computer vision, often hindered by the cost and scarcity of high-quality labeled datasets. We tackle this challenge using auxiliary datasets from related vision tasks for joint training of a shared decoder on top of a pre-trained vision foundation model, while giving a higher weight to MDE.In particular, we leverage a frozen DINOv2 ViT Giant model as a feature extractor, bypassing the need for fine-tuning, and jointly train a shared DPT decoder with auxiliary datasets from related tasks to improve MDE. We illustrate the qualitative and quantitative improvements of our method over the DINOv2 MDE baseline in Figures 1 and 2, respectively.Notably, compared to the recent Depth Anything, which reports no improvements using a jointly fine-tuned DINOv2 ViT Large and task-specific decoders, our method successfully leverages auxiliary tasks.Through extensive experiments we demonstrate the benefits of incorporating various auxiliary datasets and tasks to improve MDE quality on average by ~11% for related datasets. Our experimental analysis shows that auxiliary tasks have different impacts, confirming the importance of task selection, highlighting that quality gains are not achieved by merely adding data. Remarkably, our study reveals that using semantic segmentation datasets as multi-label dense classification often results in additional quality gains.
001034963 536__ $$0G:(DE-HGF)POF4-5111$$a5111 - Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x0
001034963 536__ $$0G:(DE-HGF)POF4-5112$$a5112 - Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x1
001034963 536__ $$0G:(DE-Juel1)Helmholtz-SLNS$$aSLNS - SimLab Neuroscience (Helmholtz-SLNS)$$cHelmholtz-SLNS$$x2
001034963 7001_ $$0P:(DE-Juel1)191034$$aYildiz, Erenus$$b1$$ufzj
001034963 7001_ $$0P:(DE-Juel1)199019$$aCao, Zhuo$$b2$$ufzj
001034963 7001_ $$0P:(DE-Juel1)151166$$aMorrison, Abigail$$b3$$ufzj
001034963 7001_ $$0P:(DE-Juel1)129347$$aKrajsek, Kai$$b4$$ufzj
001034963 7001_ $$0P:(DE-Juel1)188313$$aAssent, Ira$$b5$$ufzj
001034963 7001_ $$0P:(DE-Juel1)129394$$aScharr, Hanno$$b6$$ufzj
001034963 909CO $$ooai:juser.fz-juelich.de:1034963$$pVDB
001034963 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)188471$$aForschungszentrum Jülich$$b0$$kFZJ
001034963 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)191034$$aForschungszentrum Jülich$$b1$$kFZJ
001034963 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)199019$$aForschungszentrum Jülich$$b2$$kFZJ
001034963 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)151166$$aForschungszentrum Jülich$$b3$$kFZJ
001034963 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)129347$$aForschungszentrum Jülich$$b4$$kFZJ
001034963 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)188313$$aForschungszentrum Jülich$$b5$$kFZJ
001034963 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)129394$$aForschungszentrum Jülich$$b6$$kFZJ
001034963 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5111$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x0
001034963 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5112$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x1
001034963 9141_ $$y2024
001034963 920__ $$lyes
001034963 9201_ $$0I:(DE-Juel1)IAS-8-20210421$$kIAS-8$$lDatenanalyse und Maschinenlernen$$x0
001034963 9201_ $$0I:(DE-Juel1)IAS-6-20130828$$kIAS-6$$lComputational and Systems Neuroscience$$x1
001034963 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x2
001034963 980__ $$aconf
001034963 980__ $$aVDB
001034963 980__ $$aI:(DE-Juel1)IAS-8-20210421
001034963 980__ $$aI:(DE-Juel1)IAS-6-20130828
001034963 980__ $$aI:(DE-Juel1)JSC-20090406
001034963 980__ $$aUNRESTRICTED