Refine
Department, Institute
- Fachbereich Informatik (105)
- Institute of Visual Computing (IVC) (90)
- Institut für Sicherheitsforschung (ISF) (6)
- Institut für funktionale Gen-Analytik (IfGA) (6)
- Fachbereich Angewandte Naturwissenschaften (2)
- Fachbereich Elektrotechnik, Maschinenbau, Technikjournalismus (2)
- Fachbereich Wirtschaftswissenschaften (2)
Document Type
- Conference Object (76)
- Article (14)
- Report (12)
- Part of a Book (6)
Year of publication
Keywords
- FPGA (3)
- Image Processing (3)
- Virtual Reality (3)
- virtual reality (3)
- Intelligent virtual agents (2)
- Virtuelle Realität (2)
- Visualization (2)
- computer vision (2)
- machine vision (2)
- serious games (2)
In this contribution a machine vision inspection system is presented which is designed as a length measuring sensor. It is developed to be applied to a range of heat shrink tubes, varying in length, diameter and color. The challenges of this task were the precision and accuracy demands as well as the real-time applicability of the entire approach since it should be realized in regular industrial line production. In production, heat shrink tubes are cut to specific sizes from a continuous tube. A multi-measurement strategy has been developed, which measures each individual tube segment several times with sub pixel accuracy while being in the visual field of a standard video camera. The developed approach allows for a contact-free and fully automatic control of 100% of produced heat shrink tubes according to the given requirements with a measuring precision of 0.1mm. Depending on the color, length and diameter of the tubes considered, a true positive rate of 99.99% to 100% has been reached at a true negative rate of > 99.7.
Interactive Distributed Rendering of 3D Scenes on Multiple Xbox 360 Systems and Personal Computers
(2012)
In interactive visualization environments which use a multiple screen setup, every output device has to be supplied frequently with video information. Such virtual environments often use large projection screens, which require high resolution video data. When increasing scene complexity, it can become a challenge to equip a single computer with graphics hardware powerful enough for the task. An efficient approach is to distribute the workload to multiple low-cost computer systems such as game consoles. Nowadays' game consoles are very powerful and specialized for interactive graphics applications, therefore they are suitable to being used for rendering purposes. A framework has been developed that builds on Microsoft's XNA Game Studio. It enables for interactive distributed rendering on multiple Xbox 360 systems and PCs. Tasks such as game logic synchronization, network session management are fully handled by the framework. A game built with it can focus mostly on its own implementation. The framework's structure follows that of the XNA Game Studio, which allows porting existing game projects quickly. Our evaluation showed that the framework is a lightweight solution which leaves almost the full CPU time to the actual game.
Grailog embodies a systematics to visualize knowledge sources by graphical elements. Its main benefit is that the resulting visual presentations are easier to read for humans than the original symbolic source code. In this paper we introduce a methodology to handle the mapping from Datalog RuleML, serialized in XML, to an SVG representation of Grailog, also serialized in XML, via eXtensible Stylesheet Language Transformations (XSLT) 2.0/XML; the SVG is then rendered visually by modern Web browsers. This initial mapping is realized to target Grailog's "fully node copied" normal form. Elements can thus be translated one at a time, separating the fundamental Datalog-to-SVG translation concern from the concern of merging node copies for optimal (hyper)graph layout and avoiding its high computational complexity in this online tool. The resulting open source Grailog Knowledge-Source Visualizer (Grailog KS Viz) supports Datalog RuleML with positional relations of arity n>1. The on-the-fly transformation was shown to run on all recent major Web browsers and should be easy to understand, use, and extend.
In interactive multi-screen visualization environments, every output device has to be constantly supplied with video information. Such visualization environments often use large projection screens, which require high resolution visualization data. An efficient approach to master this challenge is to distribute the workload to multiple low-cost computer systems. Nowadays’ game consoles are very powerful and specialized for interactive graphics applications; therefore they are well suited to be applied for computational expensive rendering purposes in real-time applications. The proposed solution (dXNA) has been developed on Microsoft’s XNA Game Studio. It supports interactive distributed rendering on multiple Xbox 360 and PC setups. Application logic synchronization and network session management are completely handled by dXNA. The interface of dXNA is similar to XNA Game Studio’s interface, which allows for efficient porting of existing projects. It has been proven that dXNA is an efficient and lightweight solution for distributed rendering for interactive multi-screen visualization environments.
In this work, we automatically distinguish the efficient high elbow pose from dropping one in pulling phase of front crawl stroke in front view amateurly recorded videos. This task is challenging due to the aquatic environment and missing depth information. We predict the pull’s efficiency through multiclass svm and random forest classifiers given arms key positions and angles as the feature set. We evaluate our approach over a labeled dataset of video frames taken from 25 members of masters’ swim club at Ryerson University with different levels of expertise and physiological characteristics. Our results show the effectiveness of our approach with random forest classifier, yielding 67% accuracy.
Video surveillance has become a hot research topic due to the recently increased importance of safety and security issues. Usually, security personnel has to monitor a surveillance area and often they have to do this for 24 h a day. Thus, it would be desirable to develop intelligent surveillance systems that support this task automatically. The system described in this contribution is thought of such an automatic surveillance system that has been developed to detect several dangerous situations in subway stations. The workflow and the most important steps from foreground segmentation, shadow detection, tracking and classification to event detection are described, discussed and evaluated in detail. The developed surveillance system yields satisfying results, as dangerous situations that need to be recognized are detected in most cases.
This contribution presents an easy to implement 3D tracking approach that works with a single standard webcam. We describe the algorithm and show that it is well suited for being used as an intuitive interaction method in 3D video games. The algorithm can detect and distinguish multiple objects in real-time and obtain their orientation and position relative to the camera. The trackable objects are equipped with planar patterns of five visual markers. By tracking (stereo) glasses worn by the user and adjusting the in-game camera's viewing frustum accordingly, the well-known immersive "screen as a window" effect can be achieved, even without the use of any special tracking equipment.
In this contribution a machine vision inspection system is presented which is designed as a length measuring sensor. It is developed to be applied to a range of heat shrink tubes, varying in length, diameter and color. The challenges of this task were the precision and accuracy demands as well as the real-time applicability of the entire approach since it should be realized in regular industrial line production. In production, heat shrink tubes are cut to specific sizes from a continuous tube. A multi-measurement strategy has been developed, which measures each individual tube segment several times with sub pixel accuracy while being in the visual field. The developed approach allows for a contact-free and fully automatic control of 100% of produced heat shrink tubes according to the given requirements with a measuring precision of 0.1mm. Depending on the color, length and diameter of the tubes considered, a true positive rate of 99.99% to 100% has been reached at a true negative rate of > 99.7.
Zentrale Archivierung und verteilte Kommunikation digitaler Bilddaten in der Pneumokoniosevorsorge
(2010)
Pneumokoniose-Vorsorgeuntersuchungen erfordern die Befundung einer Röntgenthoraxaufnahme nach ILO-Staublungenklassifikation. Inzwischen werden die benötigten Aufnahmen bereits in großem Umfang digital angefertigt und kommuniziert. Hierdurch entstehen neue Anforderungen an verwendete Technik und Workflow-Mechanismen, um einen effizienten Ablauf von Untersuchung, Befundung und Dokumentation zu gewährleisten.
Im Arbeitsschutz spielt die Detektion von Fingern, Händen und Armen, insbesondere bei handbeschickten Maschinen, eine große Rolle. Gängige Schutzeinrichtungen, wie Lichtgitter, Laserscanner oder Ultraschallschutzeinrichtungen, können zwar generell Hindernisse im Gefahrenbereich von Maschinen erkennen, sind aber nicht in der Lage, zwischen Produktionsgut und Körperteilen zu unterscheiden. Die Deutsche Gesetzliche Unfallversicherung hat deshalb seit 2004 ein Forschungsvorhaben der Fachhochschule Bonn-Rhein-Sieg im Fachbereich Informatik gefördert, das zum Ziel hatte, Menschen und deren Finger und Hände durch ihre besondere Form und Farbe zu erkennen und von Produktionsgut zu unterscheiden. Als besonders kritischer Fall wurde dabei das Arbeiten an einer Tisch- und Formatkreissäge ausgewählt: Zum einen weil sich Hölzer nur schwer von Händen unterscheiden lassen, zum anderen weil an diesen Maschinen besonders viele schwere Unfällen passieren.
Traffic simulations in current open world video games and driving simulators are still limited with respect to the complexity of the behavior of simulated agents. These limitations are typically due to scarce computational resources, but also to the applied methodologies. We suggest adding cognitive components to traffic agents in order to achieve more realistic behavior, such as opting for risky actions or occasionally breaking traffic rules. To achieve this goal, we start by adding a personality profile to each agent, which is based on the “Five Factor Model” from psychology. We test our enhancement on a specific traffic scenario where simplistic behaviors would lead to a complete standstill of traffic. Our results show that the approach resolves critical situations and keeps traffic flowing.
An approach to optimal watermark detection in the DCT domain, based on the probability density function (pdf) modeling, is presented. A new detector is proposed in order to produce reliable detection of weak watermarks for various distributions of DCT image coefficients. Efficacy of the proposed detector is proved on some common attacks, as well.
3D tracking using multiple Nintendo Wii Remotes: a simple consumer hardware tracking approach
(2009)
An easy to build and cost-effective 3D tracking solution is presented, using Nintendo Wii Remotes acting as cameras. As the hardware differs from usual tracking cameras, the calibration and tracking process has to be adapted accordingly. The tracking approach described could be used for tracking the user's motions in video games based upon physical activity (sports, fighting or dancing games), allowing the player to interact with the game in a more intuitive way than by just pressing buttons.
This contribution describes an optical laser-based user interaction system designed for virtual reality (VR) environments. The project's objective is to realize a 6-DoF user input device for interaction with VR applications running in CAVE-type visualization environments with flat projections walls. In case of a back-projection VR system, in contrast to optical tracking systems, no camera has to be placed within the visualization environment. Instead, cameras observe patterns of laser beam projections from behind the screens. These patterns are emitted by a hand-held input device. The system is robust with respect to partial occlusion of the laser pattern. An inertial measurement unit is integrated into the device in order to improve robustness and precision.
A Low-Cost Based 6 DoF Head Tracker for Usability Application Studies in Virtual Environments
(2008)
Der Mutterpass wurde als wichtiges Vorsorgeinstrument für Schwangere Anfang der sechziger Jahre in Papierform eingeführt. Er wird bei 90% aller Schwangerschaften genutzt. Seit seiner Einführung im Jahre 1968 hat jedoch die Komplexität der Vorsorgeuntersuchungen zugenommen, wie auch die Begleitumstände einer Schwangerschaft häufig komplexer geworden sind. Dies war Anlass dafür, die elektronische Abbildung des Papier basierten Mutterpasses zu entwickeln, um den gewachsenen Anforderungen der medizinischen Dokumentation und Evaluation gerecht zu werden. Eine große Herausforderung bei der Konzeption und Entwicklung des elektronischen Mutterpasses war dabei die Definition eines strukturierten und maschinenlesbaren Austauschformates. Darüber hinaus mussten weltweit neue eindeutige Identifier entwickelt werden, um den Mutterpass elektronisch abzubilden. Nach der prototypischen Realisierung einer vollständigen Version wurde im Frühjahr 2008 die Pilotierung in der Metropolregion Rhein-Neckar begonnen.
Sensitivity to many visual stimuli, and, in particular, image displacement, is reduced during a change in fixation (saccade) compared to when the eye is still. In these experiments, we studied the sensitivity of observers to ecologically relevant image translations of large, complex, real world scenes either during horizontal saccades or during fixation. In the first experiment, we found that such displacements were much less detectable during saccades than during fixation. Qualitatively, even when trans-saccadic scene changes were detectible, they were less salient and appeared slower than equivalent changes in the absence of a saccade. Two further experiments followed up on this observation and estimated the perceived magnitude of trans-saccadic apparent motion using a two-interval forced-choice procedure (Experiment 2) and a magnitude estimation procedure (Experiment 3). Both experiments suggest that trans-saccadic displacements were perceived as smaller than equivalent inter-saccadic displacements. We conclude that during saccades, the magnitude of the apparent motion signal is attenuated as well as its detectability.
In this contribution, we present several improvements to previous “inside-out” techniques for pointing interaction with large display systems. Fiducial markers are virtually projected from an interaction device's built-in camera onto the displays and overlaid to the display content. We reconstruct the 6-DoF camera pose by tracking these markers in real-time. For increased robustness, the marker pattern is dynamically adapted. We address display lag and high pixel response times by precisely timing image captures. Pointing locations are measured with sub-millimeter precision and camera positions with sub-centimeter precision. An update rate of 60 Hz and a latency of 24 ms were achieved. Our technique performed comparably to an OptiTrack system in 2D target selection tasks.
In this paper we present an ongoing research work dedicated to a Virtual-Reality-based product customization application development. The work is addressing the problem of flexible and quick customization of products from a great number of parts. Our application is an effective instrument that can be simultaneously used by two users for rapid assembly tasks, allowing engineers and designers to work collaboratively. Furthermore, it is directly connected to a manufacturing environment, which is able to produce the product right after customization. In the paper we describe the architecture of the application, our interaction and assembly techniques, and explain how the system can be integrated into a manufacturing environment.
Many modern computer games take place in urban environments. One important fact about such environments is that there are plenty of other people around who might react to whatever the user does. For example in the "FIVIS" project a bicycle simulator is developed that will serve (beside other applications) as an instrument for traffic education which strongly requires other traffic participants reacting to the users actions (see http://www.fivis.eu and [7]).
This contribution describes our current work on a general approach for an interactive agent system for urban environments.
Video Surveillance is in the center of research due to high importance of safety and security issues. Usually, humans have to monitor an area and often they have to do this for 24 hours a day. Thus, it would be desirable to have automatic surveillance systems that support this job automatically. The system described in this paper is such an automatic surveillance system that has been developed to detect several dangerous situations in a subway station. This paper discusses the high-level module of the system. Herein, an expert system is used to detect events.
