Auflistung nach Autor:in "Schwanecke, Ulrich"
1 - 8 von 8
Treffer pro Seite
Sortieroptionen
- TextdokumentA Dense Statistical Model of Facial Soft Tissue Thickness(INFORMATIK 2017, 2017) Gietzen, Thomas; Brylka, Robert; Schwanecke, Ulrich; Schömer, ElmarAmbient intelligence become more and more ubiquitous and help people achieving a more natural interaction with their electronically enhanced environment. One vital natural interface between humans and ambient intelligence are embodied conversational agents. Thereby, the acceptance of these virtual characters is all the greater, the more natural they look and behave. Since humans pay particular attention to the face, a natural-looking animation of the face is very important. In this paper we present a dense statistical model of facial soft tissue thickness that can be used to build accurate physics-based facial animations. The presented model not only can help to generate more natural facial animations of virtual characters but also can be used in other research domains such as forensic anthropology or medicine. Especially in the field of dentistry and orthodontics in particularly younger people and children are increasingly examined using X-ray technology. Thereby more and more volumetric images are generated, which further increase cost as well as the induced radiation dose. Here, for example, our statistical model can provide the basis for a new volumetric reconstruction process of a human’s facial bones in a cost-effective manner and with low radiation exposure.
- TextdokumentEAVE: Emotional Aerial Vehicle Evaluator(INFORMATIK 2020, 2021) Lieser, Marc; Schwanecke, Ulrich; Berdux, JörgToday, semi-autonomous quadrotors are already available at affordable prices and have the potential to become part of everyday life due to the variety of possible applications. To ensure that people feel safe around quadrotors and to optimize flight times, their size should be kept to a minimum which results in their appearances remaining purely functional. This reduces the possibility of adding anthropomorphic or zoomorphic features that are typically used in order to increase acceptability by conveying the robot's intent or emotion. Constrained by mechanical appearance, other non-verbal communication channels can be exploited instead, in particular robot motion. The application EAVE presented in this paper was developed with the idea to design and evaluate trajectories that breathe life into inanimate, mechanical quadrotors in order to improve interaction in human-robot companionships. It extends our existing quadrotor testbed ICARUS, which is capable of tracking arbitrary trajectories of real and simulated quadrotors that were designed using EAVE. We demonstrate that applying some of the established principles of character animation to the design of quadrotor trajectories opens up the possibility of conveying intent and improving interaction, though the appearance of the quadrotor remains purely functional.
- KonferenzbeitragEvaluating Contextualized Code Search in Practical User Studies(INFORMATIK 2024, 2024) Villmow, Johannes; Ulges, Adrian; Schwanecke, UlrichContextualized Code Search (CCS) aims to retrieve relevant code snippets that complement the developer’s current editor context. In contrast to AI-based code generation, it offers the key benefit that the source of the retrieved code is made transparent, allowing for a safe re-use of code within companies. Recently, self-supervised training for CCS has been shown to be effective. Evidence for this, however, focuses on ranking quality on research datasets. It remains unclear whether – and if yes, by how far – CCS can help improve the efficiency of real-world users. To fill this gap, we have integrated a recent CCS model into an IDE. We describe specialized robustness-oriented enhancements to the training to improve usability. We then evaluate the model in two practical user studies: In Study A, we measure efficiency improvements of fourth semester computer science students on simple algorithm exercises. In Study B, we allow a professional software development team to use the tool in their everyday work. Their company consists of several – more or less independent – teams that work on the same product, which might find code of other teams helpful. We demonstrate improvements by the proposed search, discuss use cases for the tool, and point out challenges and directions for future research (such as the combination with code generation in retrieval augmented generation).
- KonferenzbeitragGoal-snapping: an empirical evaluation of object snapping in tangible and multi-touch interfaces(ARCS 2012 Workshops, 2012) Schmitt, Sebastian; Luderschmidt, Johannes; Haubner, Nadia; Lehmann, Simon; Dörner, Ralf; Schwanecke, UlrichWe present “Goal-snapping”, a novel approach for applying snapping techniques to tangible and multi-touch interfaces. It can be used to support users in accomplishing basic tasks such as aligning, sorting or grouping of virtual objects. As using snapping on large surfaces poses challenges in interaction design, we identify and discuss according parameters in Goal-snapping. For sorting and aligning, we propose to use snappers that attract objects within a target zone and visually arrange them to present an overview. For exchanging objects among users, we propose that each user has a target snapper that acts as a goal to which objects can be flicked. A user study has shown that although participants embrace the use of snapping to automatically group objects in a sorting task, snapping does not accelerate the completion time and increases the error rate by accidently snapped objects. In a long distance positioning task, the use of snapping significantly increases task completion.
- KonferenzbeitragInteractive high definition 3D face rendering on common mobile devices(Mensch und Computer 2008: Viel Mehr Interaktion, 2008) Barth, Peter; Schwanecke, Ulrich; Wild, FriederikeSmartphones are suitable as interactive visualization platform for high quality 3D models representing, for example, faces. We provide an efficient implementation of a point-based renderer in Java. A 3D-model is adequately reduced in size, represented as memory efficient octree and converted to Java code for packaging and deployment on smartphones supporting Java Mobile Edition (JavaME). The render algorithm is carefully implemented in platform independent Java and performs well on common mobile devices. While interacting with the model, we routinely achieve more than 10 frames per second providing for an agile user experience, while sustaining perceived high quality by the user.
- KonferenzbeitragTowards a top-view detection of body parts in an interactive tabletop environment(ARCS 2012 Workshops, 2012) Haubner, Nadia; Schwanecke, Ulrich; Dörner, Ralf; Lehmann, Simon; Luderschmidt, JohannesIntegrating digital tabletop systems in private living environments is a promising approach to enhance people's everyday life with information technology. Apart from using the surface of such a tabletop, research on the detection of interaction above and around the surface is increasing rapidly. So far, detection is limited either to very specific gestures above the surface or to rather abstract detection of users in a larger scenario. The detection of body parts in tabletop setups has rarely been investigated, although the knowledge about the whereabouts of body parts would be helpful to establish relationships between users and interactions. In this paper, we propose a system that is capable to detect body parts above and around such a tabletop setup using a depth camera. We further take up an existing approach to present how the detection in this setup could work. Additionally, we propose a new approach to obtain training data for the detection using a color suit.
- KonferenzbeitragTowards integration of user interaction and context event processing in intelligent living environments(ARCS 2012 Workshops, 2012) Lehmann, Simon; Schäfer, Jan; Dörner, Ralf; Schwanecke, UlrichEvent processing plays a significant role in the current development of intelligent living environments. It ranges from processing of information produced by a magnitude of sensors to gain insight into the activities of the inhabitants on a more global scale, to the processing of immediate and rather short-lived events of user input on and around interactive systems embedded in common household furniture like tabletops or tablets. Based on the work conducted separately in those two fields, we found that the still evolving field of complex event processing (CEP) provides the methods and tools to handle those distinct use-cases equally. Especially the application to interactive systems, while being novel and uncommon, is well suited and further shows the broad applicability of CEP. The comparison of the two application fields shows that, even though the events occurring in them are distinguished by their intention, commonalities do exist and provide integration points. Furthermore, the integration of those applications within the context of smart homes allows to provide demand-oriented resource management, which realizes self adaptation and control.
- TextdokumentVCIT - Visually Corrected inertial Tracking(INFORMATIK 2017, 2017) Andrés López, Daniel; Diensberg, Benedikt; Schömer, Elmar; Schwanecke, UlrichMany smart systems depend on exact models of their environment. These are gained by tracking objects in their surroundings. When a highly precise system is only available for a small part of the environment, it can be enhanced with a second system to recover the unknown parts. This paper presents a method to recover loss of a precise (optical) tracking system by a less precise (inertial) tracking system. First the rotation from the inertial measurement unit (IMU) and the optical system are aligned. A second step integrates the IMU acceleration two times and removes both times the drift by known initial and end values (first integration: velocity, second integration: position) from optical tracking. The error is backpropagated continuously.