Seite - iii - in Joint Austrian Computer Vision and Robotics Workshop 2020

TheProblemofFragmentedOcclusion inObjectDetection JulianPegoraroandRomanPflugfelder . . . . . . . . . . . . . . . . . . . . . . . . 98 ACenterline-GuidedApproach forAortaandStent-GraftSegmentation Bertram Sabrowsky-Hirsch, Stefan Thumfart, Richard Hofer, Wolfgang Fenz, Pierre Schmit, andFranzFellner . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 ImageSynthesis inSO(3)byLearningEquivariantFeatureSpaces MarcoPeer,StefanThalhammer,andMarkusVincze . . . . . . . . . . . . . . . . . 108 FrameBorderDetection forDigitizedHistoricalFootage DanielHelm,BernhardPointner, andMartinKampel . . . . . . . . . . . . . . . . . 114 HighlyAccurateBinary ImageSegmentation forCars ThomasHeitzingerandMartinKampel . . . . . . . . . . . . . . . . . . . . . . . . 116 PowderBedAnalysis inAdditiveManufacturingUsing ImageProcessing FlorianReclaandMartinWelk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 GraspingPointPrediction inClutteredEnvironmentusingAutomaticallyLabeledData StefanAinetterandFriedrichFraundorfer . . . . . . . . . . . . . . . . . . . . . . . 124 TheDifficultiesofDetectingDeformableObjectsUsingDeepNeuralNetworks NikolaDjukic,WalterG.Kropatsch,andMarkusVincze . . . . . . . . . . . . . . . . 131 BorderPropagation: ANovelApproachToDetermineSlopeRegionDecompositions FlorianBogner,AlexanderPalmrich,andWalterG.Kropatsch . . . . . . . . . . . . 137 HowHigh is theTide? EstimationofFloodLevel fromSocialMedia Julia Strebl, Djordje Slijepcevic, Armin Kirchknopf, Muntaha Sakeena, and Markus Seidl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 Real-WorldVideoRestorationusingNoise2Noise MartinZachandErichKobler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 AsymptoticAnalysisofBivariateHalf-SpaceMedianFiltering MartinWelk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 360◦monitoring for robotsusingTime-of-Flight sensors ThomasMaierandBirgitHasenberger . . . . . . . . . . . . . . . . . . . . . . . . . 157 Towards Identificationof IncorrectlySegmentedOCTScans VerenaRennerandJirˇı´Hladu˚vka . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 EvaluatingCounterMeasuresagainstSIFTKeypointForensics MuhammadSalmanandAndreasUhl . . . . . . . . . . . . . . . . . . . . . . . . . 166 AutomatedGenerationof3DGarments inDifferentSizes fromaSingleScan StefanHauswiesnerandPhilippGrasmug . . . . . . . . . . . . . . . . . . . . . . . 172