SciTePress - Affiliation Profile

Universität Koblenz, Germany

Context (please use complete words only)

Field

Sort by

9 Papers

Deep Learning-Tuned Adaptive Inertia Weight in Particle Swarm Optimization for Medical Image Registration

Topics: Image Registration; Machine Learning Technologies for Vision; Medical Image Applications

Authors: Katharina Krämer ¹ , Stefan Müller ¹ and Michael Kosterhon ²

Affiliations: ¹ Institute of Computervisualistics, University of Koblenz, Universitätsstraße 1, 56070 Koblenz, Germany , ² Medical Center Johannes Gutenberg University, Langenbeckstraße 1, 55131 Mainz, Germany

Digital Object Identifier: 10.5220/0013122000003912

Proceedings of VISAPP 2025, ISBN: 978-989-758-728-3

Quick Abstract

A novel parameter training approach for Adaptive Inertia Weight Particle Swarm Optimization (AIW-PSO) using Deep Learning is proposed. In PSO, balancing exploration and exploitation is crucial, with inertia governing parameter space sampling. This work presents a method for training transfer function parameters that adjust the inertia weight based on a particle’s individual search ability (ISA) in each dimension. A neural network is used to train the parameters of this transfer function, which then maps the ISA value to a new inertia weight. During inference, the best possible success ratio and lowest average error are used as network inputs to predict optimal parameters. Interestingly, the parameters across different objective functions are similar and assume values that may appear spatially implausible, yet outperform all other considered value expressions. We evaluate the proposed method Deep Learning-Tuned Adaptive Inertia Weight (TAIW) against three inertia strategies: Constant Inertia Strategy (CIS), Linear Decreasing Inertia (LDI), Adaptive Inertia Weight (AIW) on three benchmark functions. Additionally, we apply these PSO inertia strategies to medical image registration, utilizing digitally reconstructed radiographs (DRRs). The results show promising improvements in alignment accuracy using TAIW. Finally, we introduce a metric that assesses search effectiveness based on multidimensional search space volumes.

Oasis: A Real-Time Hydraulic and Aeolian Erosion Simulation with Dynamic Vegetation

Topics: Modeling of Natural Scenes and Phenomena; Real-Time Visual Simulation; Real-Time Graphics; Point-Based Rendering; High-Performance Computing and Parallel Rendering

Authors: Alexander Nilles , Lars Günther and Stefan Müller

Affiliation: Institute for Computational Visualistics, University of Koblenz, Universitätsstr. 1, Koblenz, Germany

Digital Object Identifier: 10.5220/0013112100003912

Proceedings of GRAPP 2025, ISBN: 978-989-758-728-3

Quick Abstract

We present a novel real-time combined simulation for aeolian erosion, hydraulic erosion and vegetation, capable of transforming barren deserts with sand dunes into lush forest landscapes and vice versa using simple user interaction. Existing aeolian and hydraulic erosion methods are extended and unified using a moisture model on a layered heightmap, supporting bedrock, soil and sand as terrain materials. Vegetation uses a 3D radius-based model and is efficiently rasterized to a 2D density map via a split-Gaussian model, inhibiting erosion. Abiotic factors such as moisture, terrain slope, surface water and illumination are considered in vegetation growth and vegetation can spread radially as well as with the wind. Each plant considers the position and size of all neighboring plants as a biotic growth factor, made possible through a set of uniform grids of varying resolutions. The user can freely model different plant species by defining their ecological niche and adaptability to changes in terrain elevation and competition with other plants. Even underwater vegetation is possible. Interspecies competition can be defined freely using a competition matrix. The resulting method runs in real-time at a terrain resolution of 20482 with 2,000,000.00 plants.

Real-Time Desertscapes Simulation with CUDA

Topics: Real-Time Rendering; Modeling of Natural Scenes and Phenomena; Real-Time Visual Simulation; Real-Time Graphics; Physics-Based Modeling; Dynamic Simulation of Physical Behaviour

Authors: Alexander Nilles , Lars Günther and Stefan Müller

Affiliation: Institute for Computational Visualistics, University of Koblenz, Universitätsstr. 1, Koblenz, Germany

Digital Object Identifier: 10.5220/0012315600003660

Proceedings of GRAPP 2024, ISBN: 978-989-758-679-8

Quick Abstract

We propose a new GPU-based method capable of simulating dune formation, propagation and aeolian sand transport, based on the Desertscape Simulation model. Our method improves upon the original as well as an existing real-time GPU implementation by introducing bilinear interpolation and removing randomness, which leads to a more robust and noise-free method. We implement our method with CUDA and use atomic adds on floats to remove a previous limitation that required block-based discretization of elevation values. A new sand distribution scheme for sand avalanching is proposed that converges faster than the previous work. We propose and evaluate a new method for reptation, which was previously neglected. Our method further improves on the performance compared with the previous real-time method and can generate results that more closely resemble dune evolution with a bidirectional wind scheme as predicted by an accurate offline method. Our method can generate detailed, physically plausible desert environments very quickly, with possible applications in both computer graphics as well as geomorphology. With some restrictions, our method could even be used during gameplay.

Combining Two-level Data Structures and Line Space Precomputations to Accelerate Indirect Illumination

Topics: Rendering Algorithms; High-Performance Computing and Parallel Rendering

Authors: K. Keul , T. Koß , F. L. Schröder and S. Müller

Affiliation: Institute for Computational Visualistics, University of Koblenz-Landau and Germany

Digital Object Identifier: 10.5220/0007345902280235

Proceedings of GRAPP 2019/VISIGRAPP, ISBN: 978-989-758-354-4

Quick Abstract

We present a method for combining two-level data structures and directional precomputation based on the Line Space (LS). While previous work has shown that LS precomputation significantly improves ray traversal performance of typical spatial data structures, it suffers from high memory consumption and low image quality due to internal approximations. Our method combines this technique with two-level BVHs, where the LS is integrated within second-level object BVHs. The advantages are, among others, optimizations in terms of approximation accuracy and required memory. In addition, we propose a method to use an accurate BVH in combination with the fast approximation-based LS for path tracing in order to further reduce image errors of the LS while still benefitting from its gain in performance.

Related Ontology Subjects/Areas/Topics (4)

Computer Vision, Visualization and Computer Graphics		Rendering		Rendering Algorithms
High-Performance Computing and Parallel Rendering

Feature Evaluation and Management for Camera Pose Tracking on 3D Models

Topics: Image Registration; Features Extraction; Tracking and Visual Navigation

Authors: Martin Schumann , Jan Hoppenheit and Stefan Müller

Affiliation: University of Koblenz, Germany

Digital Object Identifier: 10.5220/0004685905620569

Proceedings of VISAPP 2014/VISIGRAPP, ISBN: 978-989-758-009-3

Quick Abstract

Our tracking approach uses feature evaluation and management to estimate the camera pose on the camera image and a given geometric model. The aim is to gain a minimal but qualitative set of 2D image line and 3D model edge correspondences to improve accuracy and computation time. Reducing the amount of feature data makes it possible to use any complex model for tracking. Additionally, the presence of a 3D model delivers useful information to predict reliable features which can be matched in the camera image with high probability avoiding possible false matches. Therefore, a quality measure is defined to evaluate and select features best fitted for tracking upon criteria from rendering process and knowledge about the environment like geometry and topology, perspective projection, light and matching success feedback. We test the feature management to analyze the importance and influence of each quality criterion on the tracking and to find an optimal weighting.

Related Ontology Subjects/Areas/Topics (6)

Computer Vision, Visualization and Computer Graphics		Image and Video Analysis		Motion, Tracking and Stereo Vision
Features Extraction		Image Registration		Tracking and Visual Navigation

MOBILE APPLICATION SUPPORTED LIFESTYLE INTERVENTION TO LOWER PROSTATE CANCER RISK

Topics: Wearable Health Informatics; Evaluation and Use of Healthcare IT; e-Health; Development of Assistive Technology; Pervasive Health Systems and Services

Authors: Luuk P. A. Simons ¹ and J. Felix Hampe ²

Affiliations: ¹ Delft University of Technology, Netherlands , ² Koblenz University, Germany

Digital Object Identifier: 10.5220/0003872704430448

Proceedings of HEALTHINF 2012/BIOSTEC, ISBN: 978-989-8425-88-1

Quick Abstract

There is an existing Health (e)Coach Solution for supporting intensive lifestyle changes, which may help reduce cancer progression risks in men with low grade prostate cancer. An important challenge is to support and motivate healthy behaviors for the long term (4 years). Smart phones are increasingly used also in this age group. And mobile apps offer significant opportunities for personalized health behavior monitoring and support. Our overall proposition is that an mApps suite, as an extension to an existing online dashboard with automated emails and interpersonal coach sessions, appears promising for improving long term health behavior support. That is, if three conditions are met: 1) Using best of breed mApps from the market (benefits like ease-of-use, continuity, customer support); 2) creating easy and meaningful score conversions (from mApps to personal dashboard); 3) using automated emails as glue between three key aspects personal progress/dashboarding, user attention and motivation, mApp usage.As methodological approach, this paper focuses on two phases from a design research cycle: requirements analysis and a preliminary design solution exploration. In the analysis phase, two questions are addressed: What are effective lifestyle intervention components according to literature? What are solution properties that may enhance motivation and improve health behaviors? In the solution exploration phase our design question is how a proposed mApp extension may likely add value to the existing Health (e)Coach solution.

Related Ontology Subjects/Areas/Topics (13)

Biomedical Engineering	e-Health	Pervasive Health Systems and Services
Biomedical Signal Processing	Evaluation and Use of Healthcare IT	Physiological Computing Systems
Cloud Computing	Health Information Systems	Platforms and Applications
Development of Assistive Technology	Human-Computer Interaction	Wearable Sensors and Systems
Devices

PARAMETER AND CONFIGURATION ANALYSIS FOR NON-LINEAR POSE ESTIMATION WITH POINTS AND LINES

Topics: Active and Robot Vision; Tracking and Visual Navigation

Authors: Martin Schumann , Bernhard Reinert and Stefan Mueller

Affiliation: University of Koblenz, Germany

Digital Object Identifier: 10.5220/0003827402710276

Proceedings of VISAPP 2012/VISIGRAPP, ISBN: 978-989-8565-04-4

Quick Abstract

In markerless model-based tracking approaches image features as points or straight lines are used to estimate the pose. We introduce an analysis of parametrizations of the pose data as well as of error measurements between 2D image features and 3D model data. Further, we give a review of critical geometrical configurations as they can appear on the input data. From these results the best parameter choice for a non-linear pose estimator is proposed that is optimal by construction to handle a combined input of feature correspondences and works on an arbitrary number and choice of feature type. It uses the knowledge of the 3D model to analyze the input data for critical geometrical configurations.

Related Ontology Subjects/Areas/Topics (4)

Active and Robot Vision		Motion, Tracking and Stereo Vision		Tracking and Visual Navigation
Computer Vision, Visualization and Computer Graphics

TELARSCOPE: A GAMEBASED TOURISTIC AUGMENTED REALITY TELESCOPE

Topics: Augmented, Mixed and Virtual Environments; Virtual Tours; Hardware Technologies for Augmented, Mixed and Virtual Environments; Non-Desktop Interfaces

Authors: Dominik Grüntjens , Daniel Arndt , Jenny Beschorner , Jonas Dietterle , Jens Epe , Paul Kobold , Tobias Stüttem and Stefan Müller

Affiliation: University of Koblenz, Germany

Digital Object Identifier: 10.5220/0003827304970500

Proceedings of GRAPP 2012/VISIGRAPP, ISBN: 978-989-8565-02-0

Quick Abstract

TelARscope is a gamebased touristic augmented reality telescope. Its interaction and exploration concept is sufficient for a heterogeneous user group such as tourists. It enables not only to watch augmented content overlayed at all times statically but also to explore it completely interactively. We evaluated the telescope with 91 real tourists.

Related Ontology Subjects/Areas/Topics (6)

Augmented, Mixed and Virtual Environments		Hardware Technologies for Augmented, Mixed and Virtual Environments		Non-Desktop Interfaces
Computer Vision, Visualization and Computer Graphics		Interactive Environments		Virtual Tours

SPATIAL COLOR CONFIDENCE FOR PHYSICALLY BASED RENDERING SETTINGS ON LC DISPLAYS

Topics: Systems and Software Architectures for Rendering; Lighting and Appearance

Authors: Jörg Koch , Niklas Henrich and Stefan Müller

Affiliation: University of Koblenz-Landau, Germany

Digital Object Identifier: 10.5220/0002835001730180

Proceedings of GRAPP 2010/VISIGRAPP, ISBN: 978-989-674-026-9

Quick Abstract

Color confidence is crucial regarding physically-based rendering settings in order to produce most promising visual results. Display characterization and in particular spatial inhomogeneity correction is often neglected in physically-based rendering applications, yet, are important to achieve color confidence. By evaluating relevant display characteristics, this paper recommends a strategy for selecting the most suitable characterization model for a given device. We indicate the importance of correcting spatial inhomogeneity and, on that account, provide an extension to the applied characterization models. All characterization models as well as our proposed extensions are implemented using modern graphics hardware, therefore, applicable to real time applications. The focus is on finding an optimal characterization model which can achieve color confidence across the display while reducing characterization time and effort. All models are created using a common, single point, consumer measurement device and applied to two LC displays.

Related Ontology Subjects/Areas/Topics (4)

Computer Vision, Visualization and Computer Graphics		Rendering		Systems and Software Architectures for Rendering
Lighting and Appearance

Topic(s):

Title:
ISBN:
Year:
Acronym:
Subject:

Title:
Author:
Affiliation:
Subject:

Name:
Affiliation:
Country:
Conference:
Subject:

Name:
Country:
Subject:

Affiliation

Filter listed results, by