Liste der Fachpublikationen

Show publication details

Leveraging Public Participation in Urban Planning with 3D Web Technology

2014

Polys, Nicholas F. (General Chair) et al.: Proceedings Web3D 2014 : 19th International Conference on 3D Web Technology. New York: ACM, 2014, pp. 117-124

International Conference on 3D Web Technology (WEB3D) <19, 2014, Vancouver, BC, Canada>

In this paper we present a web-based platform that makes use of HTML5 technology and WebGL to facilitate public participation in urban planning. It consists of components that enable city administrations to present urban plans to the public and to engage with stakeholders. One of these components uses the open source library X3DOM to visualise 3D content-for example, a city model containing a 3D representation of a planned building. Since X3DOM does not need additional software to be installed on the user's system our implementation is more portable than previous work. Our solution is based on the open source software Liferay which allows it to be configured for various urban planning projects. In order to enable communication between different web application components residing in inline frames (iframes) we implemented a special message bus based on HTML5 postMessage. In this paper we describe implementation details, but we also intensively discuss the possibilities of modern web technology for urban planning. We motivate the use of such technology through three examples that can be implemented using our web application. In this paper we also present results from evaluating our application in user workshops carried out within the project urbanAPI that is funded by the European Commission. Finally, we draw conclusions and discuss possibilities for future urban planning use cases.

Show publication details

Liver Segmentation in Contrast Enhanced MR Datasets Using a Probabilistic Active Shape and Appearance Model

2014

IEEE Computer Society: IEEE 27th International Symposium on Computer-Based Medical Systems : CBMS 2014. Los Alamitos, Calif.: IEEE Computer Society Conference Publishing Services (CPS), 2014, pp. 523-524

IEEE Symposium on Computer-Based Medical Systems (CBMS) <27, 2014, New York, NY, USA>

The current standard for diagnosing liver tumors is contrast-enhanced multiphase computed tomography. On this basis, several software tools have been developed by different research groups worldwide to support physicians for example in measuring remnant liver volume, analyzing tumors, and planning resections. Several algorithms have been developed to perform these tasks. Most of the time, the segmentation of the liver is at the beginning of the processing chain. Therefore, a vast amount of CT-based liver segmentation algorithms have been developed. However, clinics slowly move from CT as the current gold standard for diagnosing liver diseases towards magnetic resonance imaging. In this work, we utilize a Probabilistic Active Shape Model with an MR specific preprocessing and appearance model to segment the liver in contrast enhanced MR images. Evaluation is based on 8 clinical datasets.

Show publication details
Bieber, Gerald; Kirste, Thomas; Gäde, Michael

Low Sampling Rate for Physical Activity Recognition

2014

Proceedings of the 7th International Conference on PErvasive Technologies Related to Assistive Environments : PETRA 2014. New York: ACM, 2014, Article 15, 8 p.

ACM International Conference on PErvasive Technologies Related to Assistive Environments (PETRA) <7, 2014, Rhodes Island, Greece>

The monitoring of physical activity by acceleration sensors is very common. Smartphones and it´s accessories (Smartwatch, wrist bands) are equipped with sensors and provide enough calculation power for data processing. Body worn mobile devices are recognizing various types of physical activities. The current concept consists of a very high sampling rate, the higher the sampling rate, the better the accuracy of classification. This strategy reduces the battery lifetime, especially for devices with limited physical dimensions, e.g. Smartwatches. Since sampling rate is a relevant factor for energy consumption, this work is analyzing the possibilities and performance of a very low sampling rate for physical activity recognition on Smartwatches. This work proposes the new concept of extremely low sampling rate for physical activity recognition.

Show publication details
Lukas, Uwe von; Vahl, Matthias; Mesing, Benjamin

Maritime Applications of Augmented Reality - Experiences and Challenges

2014

Shumaker, Randall (Ed.) et al.: Virtual Augmented and Mixed Reality. Applications of Virtual and Augmented Reality. Proceedings Part II : VAMR 2014. Berlin, Heidelberg, New York: Springer, 2014. (Lecture Notes in Computer Science (LNCS) 8526), pp. 465-475

International Conference Virtual Augmented and Mixed Reality (VAMR) <6, 2014, Heraklion, Crete, Greece>

The paper summarizes experiences from applied research in visual computing for the maritime sector. It starts with initial remarks on Augmented Reality in general and the specific boundary conditions of the maritime industry. The focus is on a presentation of various concrete AR applications that have been implemented for use cases in maritime engineering, production, operation and retrofitting. The paper closes with remarks on future research in this area.

Show publication details
Puhl, Julian; Kuijper, Arjan (Betreuer); Knuth, Martin (Betreuer)

Materialsysteme für das realistische Echtzeit-Rendering von Szenen in Anwesenheit von Flüssigkeitssimulationen und Image-Based Lighting

2014

Darmstadt, TU, Bachelor Thesis, 2014

Für eine Simulation einer Flüssigkeit in Echtzeit werden aktuell Verfahren angewandt, die entweder das Modell diskretisieren und somit gitterbasiert arbeiten oder die einzelnen Atome zu größeren Partikeln zusammenfassen und so das Verhalten simulieren. Beiden Ansätzen ist gemein, dass als Ergebnis keine glatte Oberfläche vorliegt, sondern eine angenäherte, die uneben ist, aus vielen einzelnen Elementen besteht und zudem keine optischen Eigenschaften einer Flüssigkeit besitzt. Aus Zeitgründen darf die nachfolgende Generierung der Oberfläche nicht sehr zeitaufwendig sein. Aus diesem Grund wird in dieser Arbeit die Nachbearbeitung der Daten via Bildverarbeitung unter der Anwendung von Glättungsfiltern sowie einer Bildpyramide untersucht. Die Pyramide bietet Zugriff auf verschiedene Detailstufen. Hierdurch können unterschiedliche Glättungen in Abhängigkeit von der Entfernung zum Betrachter gewählt werden. Viele Verfahren filtern hier sehr stark und fügen nachträglich wieder ein Rauschen ein, um eine nicht ganz so glatte Oberfläche zu simulieren. In Szenen, bei denen die simulierte Flüssigkeit seitlich betrachtet wird, können sowohl nahe als auch weiter entfernte Partikel nahe beieinander existieren. Hier spielt das Verfahren seine Stärke aus gleichzeitig auf unterschiedlich stark gefilterte Werte zugreifen zu können.

Show publication details
Dekker, Niels; Lucassen, Marcel; Kirchner, Eric; Urban, Philipp; Huertas, Huertas

Mathematical Limitations when Choosing Psychophysical Methods: Geometric Versus Linear Grey Scales

2014

Segovia, Maria V. Ortiz (Ed.) et al.: Measuring, Modeling, and Reproducing Material Appearance. Bellingham: SPIE Press, 2014. (Proceedings of SPIE 9018), pp. 90180G-1 - 90180G-11

Measuring, Modeling, and Reproducing Material Appearance (MMRMA) <1, 2014, San Francisco, CA, USA>

The grey scale method is commonly used for investigating differences in material appearance. Specifically, for testing color difference equations, perceived color differences between sample pairs are obtained by visually comparing to differences in a series of achromatic sample pairs. Two types of grey scales are known: linear and geometric. Their instrumental color differences vary linearly or geometrically (i.e., exponentially), respectively. Geometric grey scales are used in ISO standards and standard procedures of the textile industries.

Show publication details

Measuring Context Relevance for Adaptive Semantics Visualizations

2014

Lindstaedt, Stefanie (Ed.) et al.: i-KNOW 2014 : Proceedings of the 14th International Conference on Knowledge Technologies and Data-driven Business. New York: ACM, 2014. (ACM International Conference Proceedings Series 889), Article 14, 8 p.

International Conference on Knowledge Technologies and Data-driven Business (I-KNOW) <14, 2014, Graz, Austria>

Semantics visualizations enable the acquisition of information to amplify the acquisition of knowledge. The dramatic increase of semantics in form of Linked Data and Linked-Open Data yield search databases that allow to visualize the entire context of search results. The visualization of this semantic context enables one to gather more information at once, but the complex structures may as well confuse and frustrate users. To overcome the problems, adaptive visualizations already provide some useful methods to adapt the visualization on users' demands and skills. Although these methods are very promising, these systems do not investigate the relevance of semantic neighboring entities that commonly build most information value. We introduce two new measurements for the relevance of neighboring entities: The Inverse Instance Frequency allows weighting the relevance of semantic concepts based on the number of their instances. The Direct Relation Frequency inverse Relations Frequency measures the relevance of neighboring instances by the type of semantic relations. Both measurements provide a weighting of neighboring entities of a selected semantic instance, and enable an adaptation of retinal variables for the visualized graph. The algorithms can easily be integrated into adaptive visualizations and enhance them with the relevance measurement of neighboring semantic entities. We give a detailed description of the algorithms to enable a replication for the adaptive and semantics visualization community. With our method, one can now easily derive the relevance of neighboring semantic entities of selected instances, and thus gain more information at once, without confusing and frustrating users.

Show publication details
Happel, Kathrin; Dörsam, Edgar; Urban, Philipp

Measuring Isotropic Subsurface Light Transport

2014

Optics Express, Vol.22 (2014), 8, pp. 9048-9062

Subsurface light transport can affect the visual appearance of materials significantly. Measuring and modeling this phenomenon is crucial for accurately reproducing colors in printing or for rendering translucent objects on displays. In this paper, we propose an apparatus to measure subsurface light transport employing a reference material to cancel out adverse signals that may bias the results. In contrast to other approaches, the setup enables improved focusing on rough surfaces (e.g. uncoated paper). We derive a measurement equation that may be used to deduce the point spread function (PSF) of subsurface light transport. Main contributions are the usage of spectrally-narrowband exchangeable LEDs allowing spectrally-resolved measurements and an approach based on quadratic programming for reconstructing PSFs in the case of isotropic light transport.

Show publication details
Segovia, Maria V. Ortiz; Urban, Philipp; Allebach, Jan P.

Measuring, Modeling, and Reproducing Material Appearance

2014

Measuring, Modeling, and Reproducing Material Appearance (MMRMA) <1, 2014, San Francisco, CA, USA>

Proceedings of SPIE 9018

The main topics encountered on the papers of this document correspond to any of the following categories: Methods for measuring material properties: measurement of Bidirectional Reflectance Distribution Functions (BRDF), Bidirectional Texture Functions (BTF) and Bidirectional Surface Scattering Reflectance Distribution Function (BSSRDF); estimation of material difference perception; evaluation of metallic coatings/inks; measurement of glossiness; estimation of texture perception; data acquisition methods for different types of materials. Models for distinct characteristics of materials: modeling of Bidirectional Reflectance Distribution Functions (BRDF), Bidirectional Texture Functions (BTF) and Bidirectional Surface Scattering Reflectance Distribution Function (BSSRDF); modeling material difference perception; appearance modeling of glossiness and texture; modeling of varnish and special effects inks; softproofing methods for 2.5D and 3D printing. Material reproduction aspects: quality evaluation of 2.5D and 3D soft- and hardcopy reproductions (display and printing); estimation of effects of environmental aspects in material perception (lighting, observers' position, printing media); estimation of sensory input (visual, touch, audio) effect in material perception; evaluation of aesthetic aspects of 2.5D and 3D soft- and hard-copy reproductions (display and printing); saliency of 2.5D and 3D soft- and hard-copy reproductions (display and printing); imaging and perception of metallic and effect coatings/inks; saliency, quality, and aesthetics in appearance reproduction; spectral reproduction.

Show publication details

Mesh Partitioning for Parallel Garment Simulation

2014

Skala, Vaclav (Ed.): WSCG 2014. Communication Papers Proceedings : 22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision. Plzen: University of West Bohemia, 2014, pp. 125-133

International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision (WSCG) <22, 2014, Plzen, Czech Republic>

We present a method for partitioning meshes that allows a simple and efficient parallel implementation of different simulation methods. It is based on a generalization of the concept of independent sets from graph theory to sets of simulation elements. The general description makes it versatile and flexibly applicable in existing simulation systems. Every simulation method that formerly worked by sequentially processing a set of simulation elements can now be parallelized by partitioning the underlying set, without affecting the behavior of the simulated model.

Show publication details
Kopp, Manuel Georg Maria; Kuijper, Arjan (Betreuer); Bremm, Sebastian (Betreuer); Landesberger, Tatiana von (Betreuer)

Methoden und visuell interaktive Werkzeuge zur automatischen Generierung von Graphen mit geografischem Bezug

2014

Darmstadt, TU, Master Thesis, 2014

Analyse oder Simulation von Zuliefernetzwerken werden oftmals durch grafische Darstellungen unterstützt. Dafür werden die Erstellung der Lieferstrukturen sowie die GEO-Position der einzelnen Lieferanten benötigt. Die Generierung der zugehörigen Graphen mit geografischem Bezug sollte nach Möglichkeit weitestgehend automatisiert erfolgen. Eine automatisierte Verortung steht Graphengeneratoren derzeit nicht zur Verfügung. Es kann daher sehr zeitaufwändig sein, wenn die Verortung eines Knotens verändert wird. Bei den aktuellen Verfahren müssen alle mit diesem Knoten verbundenen Knoten manuell auf ihre Lokalisation überprüft und ggf. manuell angepasst werden. In dieser Arbeit wird ein Weg aufgezeigt Graphen mit geografischem Bezug zu generieren, um diese manuellen Überprüfungen und Anpassungen zu automatisieren. Um dies zu erreichen wurde eine Datenbasis aus öffentlich zugänglichen Daten generiert. Diese beinhaltet eine politisch basierte hierarchische Aufteilung der Welt, aus der generierte Knoten automatisch verortet werden. Darüber hinaus wurden Regeln definiert, um statische Beziehungseigenschaften von Knoten zu ersetzen. Auf diese Weise werden Struktur-, Knoten- oder Kanteneigenschaften des zu erstellenden Graphen mit geografischem Bezug festgelegt. Die Regeln können interaktiv unter Verwendung einer Visualisierung manipuliert werden. Dadurch ist es möglich generisch verschiedene Graphen mit geografischem Bezug ohne großen Aufwand zu erstellen. Den Abschluss bildet ein Beispielszenario, dass unter Verwendung einiger der Regeln mehrere Graphen mit geografischem Bezug generiert. Anschließend werden diese zu einem Gesamtgraphen zusammengefasst um das gesamte Szenario darzustellen.

Show publication details
Bieber, Gerald; Kirste, Thomas (Betreuer); Urban, Bodo (Betreuer); Stricker, Didier (Betreuer)

Methodik zur mobilen Erfassung körperlicher Aktivität mittels Beschleunigungssensoren

2014

Rostock, Univ., Diss., 2014

Ein ausgewogenes Verhältnis zwischen körperlicher Aktivität und gesunder Ernährung trägt zu einem gesunden Lebensstil bei. Oftmals wird jedoch ein optimales Maß an Bewegung und Ernährung nicht erreicht, so dass Übergewicht und gesundheitliche Probleme auftreten können. Eine objektive Erhebung der Bewegungssituation kann hierbei Defizite erkennen und eine Intervention unterstützen. Für die Erfassung der körperlichen Aktivität mittels Beschleunigungssensoren war es bislang üblich, entweder nur einfachste körperliche Aktivitäten zu erfassen oder spezielle Sensoren einzusetzen, die exakt am Körper auszurichten sind. Die sinnvolle Nutzung von nur einem Sensor zur Aktivitätserfassung, der beliebig orientiert getragen wird, war bisher nicht möglich, da die erhobenen Sensordaten bei Verdrehung nur eine eingeschränkte Erkennung der ausgeführten Aktivitäten zulassen. Die vorliegende Arbeit beschreibt die Methodik zur mobilen Erfassung körperlicher Aktivität mittels Beschleunigungssensoren. Hierbei wird ein neues Verfahren eingesetzt, welches die Orientierung des Sensorsystems identifiziert und die gemessenen Beschleunigungsdaten in ein neues, geeignetes Bezugssystem transformiert. Zur Bestimmung der Ausrichtung des Messsystems werden spezifische Bewegungsinformationen einer Aktivität genutzt, die eine vollständige Bestimmung der Orientierung erlauben. Durch die Transformation der Sensordaten in ein neues Bezugssystem wird es möglich, die Menge der Trainingsdaten zu reduzieren und die Erkennungsrate zu erhöhen. Die hierfür notwendigen Teilkomponenten werden als Module definiert und in einer Gesamtarchitektur zusammengefasst. Dieses Gesamtmodell basiert auf der Aktivitätserkennung mittels merkmalsbasierter Mustererkennung. In Teilmodulen werden die Messdatenvorverarbeitungen zur Signalschätzung, die Merkmalsextraktion sowie die Klassifizierung abgebildet. Das Gesamtkonzept beinhaltet die Auswahl der relevanten Merkmale sowie eines geeigneten Klassifizierers für mobile Endgeräte. Als Nachweis der Umsetzbarkeit wurde eine Implementierung auf einem handelsüblichen Mobiltelefon durchgeführt. Die Evaluation zeigt, dass die Bestimmung beliebiger Sensororientierungen und die Generierung des Bezugssystems dynamisch möglich sind und eine praxistaugliche Aktivitätserkennung umsetzbar ist. Der Technologiedemonstrator Dia- Trace liefert den Nachweis, dass auf dieser Grundlage ein mobiles Assistenzsystem sinnvoll zur Behandlung von Adipositaspatienten eingesetzt werden kann.

Show publication details
Stenin, Igor; Hansen, Stefan; Becker, Meike; Sakas, Georgios; Fellner, Dieter W.; Klenzner, Thomas; Schipper, Jörg

Minimally Invasive Multiport Surgery of the Lateral Skull Base

2014

BioMed Research International, Vol.2014 (2014), Article ID 379295, 7 p.

<b>Objective:</b> Minimally invasive procedures minimize iatrogenic tissue damage and lead to a lower complication rate and high patient satisfaction. To date only experimental minimally invasive single-port approaches to the lateral skull base have been attempted. The aim of this study was to verify the feasibility of a minimally invasive multiport approach for advanced manipulation capability and visual control and develop a software tool for preoperative planning. <b>Methods:</b> Anatomical 3D models were extracted from twenty regular temporal bone CT scans. Collision-free trajectories, targeting the internal auditory canal, round window, and petrous apex, were simulated with a specially designed planning software tool. A set of three collision-free trajectories was selected by skull base surgeons concerning the maximization of the distance to critical structures and the angles between the trajectories. <b>Results:</b> A set of three collision-free trajectories could be successfully simulated to the three targets in each temporal bone model without violating critical anatomical structures. <b>Conclusion:</b> A minimally invasive multiport approach to the lateral skull base is feasible. The developed software is the first step for preoperative planning. Further studies will focus on cadaveric and clinical translation.

Show publication details
Aehnelt, Mario; Bader, Sebastian

Mobile Informationsassistenz für die Montage

2014

Weidner, Robert (Ed.) et al.: Technische Unterstützungssysteme, die die Menschen wirklich wollen : Erste Transdisziplinäre Konferenz zum Thema. Hamburg: Helmut-Schmidt-Univ., 2014, pp. 270-280

Transdisziplinäre Konferenz zum Thema Technische Unterstützungssysteme <1, 2014, Hamburg, Germany>

Montagearbeiter zeichnen sich gegenüber automatisierten Montagesystemen nicht nur durch ihre Flexibilität sondern auch durch kreative Lösungsstrategien aus. Diese beruhen auf Erfahrungswissen und einer großen kognitiven Leistungsfähigkeit. Sie erfordern jedoch auch eine situativ angepasste Informationsassistenz. Der Beitrag beschreibt ein intelligentes System zur Informationsunterstützung des Montagearbeiters im Bereich der Kleinserien- und Einzelfertigung. Aufbauend auf vorangegangenen Arbeiten wird ein Framework aus intelligenten Komponenten, z.B. wahrscheinlichkeitsbasierten Verhaltensmodellen, kognitiven Architekturen sowie mobilen Assistenzlösungen, vorgestellt. Dieses unterstützt den Montagearbeiter in Abhängigkeit seiner Arbeitsaufgaben durch eine bedarfsgerechte Bereitstellung von Arbeitsanweisungen und es überwacht die korrekte Durchführung der Montagearbeiten.

Show publication details
Nesterenko, Dmitrijs; Aehnelt, Mario (Betreuer); Urban, Bodo (Betreuer)

Mobile und situative Informationsverteilung auf verteilten Displays

2014

Rostock, Univ., Bachelor Thesis, 2014

Bei der Montage von Maschinen und Anlagen werden Informationen über Aufträge, konstruktive oder verfahrenstechnische Details sowie zu verarbeitende Bauteile und Materialien benötigt. Zur Sicherstellung ihrer Aktualität sind digitale Technologien für die Bereitstellung und Darstellung der Informationen am stationären, teil-stationären und mobilen Arbeitsplatz erforderlich. Umgebungen, die mit den Technologien ausgestattet sind, sollen die Information so verteilen, dass die Erfassung und Verarbeitung der Informationen durch den Montagearbeiter den Montageprozess nicht zusätzlich behindert sondern unterstützt. Je nach Ausstattung und Arbeitssituation ist die Informationsverteilung unterschiedlich. Im Rahmen dieser Bachelorarbeit soll ein Konzept entwickelt werden, der die situative Verteilung der Informationen auf die verteilten Displays ermöglicht. Außerdem soll ein Prototyp für das Konzept entwickelt werden und in eine bestehende Anwendung integriert werden. Die zu implementierende Funktionalitäten für den Prototyp werden im Laufe der Arbeit eingeschränkt.

Show publication details
Hammon, Matthias; Cavallaro, Alexander; Erdt, Marius; Dankerl, Peter; Kirschner, Matthias; Drechsler, Klaus; Wesarg, Stefan; Uder, Michael; Janka, Rolf

Model-Based Pancreas Segmentation in Portal Venous Phase Contrast-Enhanced CT Images

2014

Journal of Digital Imaging, Vol.26 (2014), 6, pp. 1082-1090. First published online 08 March 2013

This study aims to automatically detect and segment the pancreas in portal venous phase contrast-enhanced computed tomography (CT) images. The institutional review board of the University of Erlangen-Nuremberg approved this study and waived the need for informed consent. Discriminative learning is used to build a pancreas tissue classifier incorporating spatial relationships between the pancreas and surrounding organs and vessels. Furthermore, discrete cosine and wavelet transforms are used to build texture features to describe local tissue appearance. Classification is used to guide a constrained statistical shape model to fit the data. The algorithm to detect and segment the pancreas was evaluated on 40 consecutive CT data that were acquired in the portal venous contrast agent phase. Manual segmentation of the pancreas was carried out by experienced radiologists and served as reference standard. Threefold cross validation was performed. The algorithm-based detection and segmentation yielded an average surface distance of 1.7 mm and an average overlap of 61.2 % compared with the reference standard. The overall runtime of the system was 20.4 min. The presented novel approach enables automatic pancreas segmentation in portal venous phase contrast-enhanced CT images which are included in almost every clinical routine abdominal CT examination. Reliable pancreatic segmentation is crucial for computer-aided detection systems and an organ-specific decision support.

Show publication details
De Stefano, Antonio; Santos, Pedro (Betreuer); Tausch, Reimar (Betreuer)

Modeling a Virtual Robotic System for Cultural Heritage Digitization

2014

Neapel, Univ., Master Thesis, 2014

The objective of this thesis is the development of an inverse kinematic solver application which allows a camera mounted onto the robotic arm to resolve occlusion in a 3D model, by reaching a position and a camera direction, given in this case by a path creator, while in the reality the pose is given from a view planning; The results will be visualized and simulated in a virtual environment which represents the CultArm3D subsystem of the CultLAb3D. This thesis is part of the Cultural Heritage Project of the Fraunhofer Institute for Visual Computing, which is concerned with the development of inexpensive automated 3D-digitization of objects of cultural value. Given that many museums lack exhibition space for their numerous exhibits and that these museums are scattered around the world, the Cultural Heritage Project seeks to provide worldwide virtual access to these exhibits.

Show publication details
Schinko, Christoph; Ullrich, Torsten; Fellner, Dieter W.

Modeling with High-level Descriptions and Low-level Details

2014

Blashki, Katherine (Ed.) et al.: Proceedings of the International Conferences on Interfaces and Human Computer Interaction 2014, Game and Entertainment Technologies 2014 and Computer Graphics, Visualization, Computer Vision and Image Processing 2014 : Part of the Multi Conference on Computer Science and Information Systems, MCCSIS 2014. IADIS Press, 2014, pp. 328-332

IADIS International Conference Computer Graphics, Visualization, Computer Vision and Image Processing (CGVCVIP) <8, 2014, Lisbon, Portugal>

Procedural modeling techniques can be used to encode a geometric shape on a high and abstract level: each class of objects and shapes is represented by one algorithm; and each artifact is one set of high-level parameters. In this paper, we use a generative object description and register it to real-world data e.g. laser scans. Afterwards, we can use the fitted procedural model to modify existing 3D shapes. The high-level description can be used to resemble real-world objects or create new ones. In this way, we can design shapes using both low-level details and high-level shape parameters at the same time.

Show publication details
Djakow, Maxim; Braun, Andreas; Marinc, Alexander

MoviBed - Sleep Analysis Using Capacitive Sensors

2014

Stephanidis, Constantine (Ed.) et al.: Universal Access in Human-Computer Interaction: Part IV : Design for All and Accessibility Practice. UAHCI 2014. Berlin, Heidelberg, New York: Springer, 2014. (Lecture Notes in Computer Science (LNCS) 8516), pp. 171-181

International Conference on Universal Access in Human-Computer Interaction (UAHCI) <8, 2014, Heraklion, Crete, Greece>

Sleep disorders are a wide-spread phenomenon that can gravely affect personal health and well-being. An individual sleep analysis is a first step in identifying unusual sleeping patterns and providing suitable means for further therapy and preventing escalation of symptoms. Typically such an analysis is an intrusive method and requires the user to stay in a sleep laboratory. In this work we present a method for detecting sleep patterns based on invisibly installed capacitive proximity sensors integrated into the bed frame. These sensors work with weak electric fields and do not disturb sleep. Using the movements of the sleeping person we are able to provide a continuous analysis of different sleep phases. The method was tested in a prototypical setup over multiple nights.

Show publication details
Damer, Naser; Opel, Alexander

Multi-biometric Score-Level Fusion and the Integration of the Neighbors Distance Ratio

2014

Aurélio, Campilho (Ed.) et al.: Image Analysis and Recognition. Proceedings Part II : 11th International Conference, ICIAR 2014. Berlin; Heidelberg; New York: Springer, 2014. (Lecture Notes in Computer Science (LNCS) 8815), pp. 85-93

International Conference on Image Analysis and Recognition (ICIAR) <11, 2014, Vilamoura, Portugal>

Multi-biometrics aims at building more accurate unified biometric decisions based on the information provided by multiple biometric sources. Information fusion is used to optimize the process of creating this unified decision. In previous works dealing with score-level multibiometric fusion, the scores of different biometric sources belonging to the comparison of interest are used to create the fused score. The novelty of this work focuses on integrating the relation of the fused scores to other comparisons within a 1:N comparison. This is performed by considering the neighbors distance ratio in the ranked comparisons set within a classification-based fusion approach. The evaluation was performed on the Biometric Scores Set BSSR1 database and the enhanced performance induced by the integration of neighbors distance ratio was clearly presented.

Show publication details

Multi-Sensory Environment Analysis and Human Activity Recognition via Wearable Technologies

2014

Bieber, Gerald (Ed.) et al.: WOAR 2014 : Proceedings of the Workshop on Sensor-based Activity Recognition. Stuttgart: Fraunhofer Verlag, 2014, pp. 2-11

Workshop on Sensor-Based Activity Recognition (WOAR) <1, 2014, Rostock, Germany>

The sensing of human activities and user surrounding environments is an essential topic in computer science. Application domains include the Ambient Assisted Living (AAL), healthcare, sports gear and military use cases. Especially mobile or wearable technologies made significant progress in the past years. The current development of Smartwatches and Smartphones which include a variety of sensors is a good example for this progress. With the increasing sensor density in unobtrusive wearable designs, new ways for complex Human Activity Recognition (HAR) and environmental sensing (ES) are opened. This paper focuses on the current state of the art in wearable sensor technologies and gives a short overview of present techniques for HAR and ES. Therefore, a classification of sensors by use case and body position is made. Furthermore, the principal challenges and issues are discussed and known solutions will be referenced. Finally, existing problems that should be addressed are pointed out.

Show publication details
Ackermann, Jens; Langguth, Fabian; Fuhrmann, Simon; Kuijper, Arjan; Goesele, Michael

Multi-View Photometric Stereo by Example

2014

3DV 2014 - Volume I : 2014 International Conference on 3D Vision. Los Alamitos, Calif.: IEEE Computer Society, 2014, pp. 259-266

International Conference on 3D Vision (3DV) <2, 2014, Tokyo, Japan>

We present a novel multi-view photometric stereo technique that recovers the surface of textureless objects with unknown BRDF and lighting. The camera and light positions are allowed to vary freely and change in each image. We exploit orientation consistency between the target and an example object to develop a consistency measure. Motivated by the fact that normals can be recovered more reliably than depth, we represent our surface as both a depth map and a normal map. These maps are jointly optimized and allow us to formulate constraints on depth that take surface orientation into account. Our technique does not require the visual hull or stereo reconstructions for bootstrapping and solely exploits image intensities without the need for radiometric camera calibration. We present results on real objects with varying degree of specularity and show that these can be used to create globally consistent models from multiple views.

Show publication details
Brunton, Alan; Bolkart, Timo; Wuhrer, Stefanie

Multilinear Wavelets: A Statistical Shape Space for Human Faces

2014

Fleet, David (Ed.) et al.: Computer Vision - ECCV 2014. Proceedings Part I : 13th European Conference on Computer Vision. Berlin, Heidelberg, New York: Springer, 2014. (Lecture Notes in Computer Science (LNCS) 8689), pp. 297-312

European Conference on Computer Vision (ECCV) <13, 2014, Zurich, Switzerland>

We present a statistical model for 3D human faces in varying expression, which decomposes the surface of the face using a wavelet transform, and learns many localized, decorrelated multilinear models on the resulting coefficients. Using this model we are able to reconstruct faces from noisy and occluded 3D face scans, and facial motion sequences. Accurate reconstruction of face shape is important for applications such as tele-presence and gaming. The localized and multi-scale nature of our model allows for recovery of fine-scale detail while retaining robustness to severe noise and occlusion, and is computationally efficient and scalable. We validate these properties experimentally on challenging data in the form of static scans and motion sequences. We show that in comparison to a global multilinear model, our model better preserves fine detail and is computationally faster, while in comparison to a localized PCA model, our model better handles variation in expression, is faster, and allows us to fix identity parameters for a given subject.

Show publication details
Aginako, Naiara; Irujo Arraiza, Juan; Cuadros, Montse; Raffaelli, Matteo; Kähm, Olga; Damer, Naser; Neto, Joao P. Neto

Multimedia Analysis of Video Sources

2014

Obaidat, Mohammad S. (Ed.) et al.: SIGMAP 2014. Proceedings of the 11th International Conference on Signal Processing and Multimedia Applications. SciTePress, 2014, pp. 346-352

International Conference on Signal Processing and Multimedia Applications (SIGMAP) <11, 2014, Vienna, Austria>

Law Enforcement Agencies (LEAs) spend increasing efforts and resources on monitoring open sources, searching for suspicious behaviours and crime clues. The task of efficiently and effectively monitoring open sources is strongly linked to the capability of automatically retrieving and analyzing multimedia data. This paper presents a multimodal analytics system, created in cooperation with European LEAs. In particular it is described how the video analytics subsystem produces a workflow of multimedia data analysis processes. After a first analysis of video files, images are extracted in order to perform image comparison, classification and face recognition. In addition, audio content is extracted to perform speaker recognition and multilingual analysis of text transcripts. The integration of multimedia analysis results allows LEAs to extract pertinent knowledge from the gathered information.

Show publication details
Trimpop, John; Urban, Bodo (Betreuer); Bieber, Gerald (Betreuer)

Multisensorische Aktivitätserkennung an Arm und Kopf

2014

Rostock, Univ., Master Thesis, 2014

Innovative Smart Devices ermöglichen eine umfassende, permanente und unaufdringliche Erkennung körperlicher Aktivität. Neben Smartphones und Smartwatches existieren auf dem Markt mittlerweile auch Smartglasses verschiedener Hersteller. Da die Human Activity Recognition mit nahe der Hüfte und am Arm getragener Sensoren bereits gut funktioniert, wird im Rahmen dieser Mastarbeit untersucht, ob eine auf Kopfsensorik basierende Aktivitätserkennung alleinstehend beziehungsweise in ein multisensorisches Netzwerk eingebettet durchführbar ist. Für dieses Vorhaben werden Konzepte zur multisensorischen Aktivitätserkennung und Sensorfusion entwickelt, die in einem umfangreichen System, bestehend aus mehreren Komponenten, implementiert werden. Das System kennzeichnet sich als Aktivitätstagebuch, welches Daten von am Kopf, am Arm und in der Hosentasche getragener Sensoren verarbeitet, auswertet und speichert. Die Ergebnisse einer ausgiebigen Evaluation unter Nutzung des Systems zeigen auf, dass am Kopf befestigte Sensoren für die Aktivitätserkennung eingesetzt werden können. Durch die Integration der Kopfposition innerhalb eines multisensorischen Netzwerkes konnten die Ergebnisse kontinuierlich verbessert werden. Zusätzliche Untersuchungen präsentieren darüber hinaus, dass verschiedene Geräte und Positionen am Kopf für die Aktivitätserkennung in Frage kommen und, dass die Kopfbewegung im Hinblick auf die körperliche Aktivität sehr charakteristisch ist.