001046025 001__ 1046025
001046025 005__ 20251124202414.0
001046025 0247_ $$2datacite_doi$$a10.34734/FZJ-2025-03665
001046025 037__ $$aFZJ-2025-03665
001046025 041__ $$aEnglish
001046025 1001_ $$0P:(DE-Juel1)208654$$aKromm, Edward$$b0$$eCorresponding author
001046025 245__ $$aData Fusion for Scene Graph Generation: Bridging Simulated and Real-World Datasets$$f2025-03-11 - 2025-10-11
001046025 260__ $$aJülich$$c2025
001046025 300__ $$a102 pages: Figures, Tables
001046025 3367_ $$2DataCite$$aOutput Types/Supervised Student Publication
001046025 3367_ $$02$$2EndNote$$aThesis
001046025 3367_ $$2BibTeX$$aMASTERSTHESIS
001046025 3367_ $$2DRIVER$$amasterThesis
001046025 3367_ $$0PUB:(DE-HGF)19$$2PUB:(DE-HGF)$$aMaster Thesis$$bmaster$$mmaster$$s1763984777_15509
001046025 3367_ $$2ORCID$$aSUPERVISED_STUDENT_PUBLICATION
001046025 502__ $$aMasterarbeit, Hochschule Coburg, 2025$$bMasterarbeit$$cHochschule Coburg$$d2025$$o2025-10-11
001046025 520__ $$aScene graph generation has emerged as a powerful tool for AI-driven visual understandingof images by not only detecting objects in an image but also predicting the relationshipsbetween them, such as car–stops at–traffic light or pedestrian–crosses–street. This capabilityis particularly important for autonomous driving, where relational context between roadusers and infrastructure plays a critical role. However, the application of scene graphgeneration in this domain is hindered by the scarcity of annotated datasets. Drivingsimulators such as CARLA provide a scalable alternative, enabling efficient data generationcompared to manual annotation. Yet models trained exclusively on simulated data oftenfail to generalize to real-world data due to the substantial domain gap between the two.This thesis addresses this challenge by proposing a novel data fusion framework thatcombines simulated and real datasets to construct autonomous driving–specific relationshipannotations and subsequently bridge the domain gap for real-world prediction. The workpresents the complete pipeline, including dataset generation in simulation, adaptationof publicly available resources, and augmentation strategies. The Relation Transformermodel is analyzed in depth, and particular attention is given to interpreting its internalmechanisms by visualizing the learned attention maps as heatmaps. This analysis providesinsights into whether the model focuses on semantically meaningful regions when predictingrelationships. Building on this understanding, two new approaches are introduced to enableinference on real data while transferring relational knowledge acquired in simulation. Anablation study further quantifies the impact of the domain gap on model performance andhighlights the strengths and limitations of the proposed methods. Results demonstratethat one of the developed approaches effectively mitigates the simulation-to-reality gapand concrete suggestions for advancing this technique toward further uses for AI-drivenvisual understanding of images in the automotive context are provided.
001046025 536__ $$0G:(DE-HGF)POF4-5112$$a5112 - Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x0
001046025 536__ $$0G:(BMWK)19A23014l$$anxtAIM - nxtAIM – NXT GEN AI Methods (19A23014l)$$c19A23014l$$x1
001046025 7001_ $$0P:(DE-Juel1)192312$$aBenassou, Sabrina$$b1$$eThesis advisor
001046025 7001_ $$0P:(DE-Juel1)185654$$aKesselheim, Stefan$$b2$$eThesis advisor
001046025 8564_ $$uhttps://juser.fz-juelich.de/record/1046025/files/Master%20Thesis.pdf$$yOpenAccess
001046025 909CO $$ooai:juser.fz-juelich.de:1046025$$popenaire$$popen_access$$pVDB$$pdriver$$pdnbdelivery
001046025 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)208654$$aForschungszentrum Jülich$$b0$$kFZJ
001046025 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)192312$$aForschungszentrum Jülich$$b1$$kFZJ
001046025 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)185654$$aForschungszentrum Jülich$$b2$$kFZJ
001046025 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$$x0
001046025 9141_ $$y2025
001046025 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
001046025 920__ $$lyes
001046025 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x0
001046025 980__ $$amaster
001046025 980__ $$aVDB
001046025 980__ $$aUNRESTRICTED
001046025 980__ $$aI:(DE-Juel1)JSC-20090406
001046025 9801_ $$aFullTexts