Welcome on the homepage of the chair "Computer Graphics Systems" of Prof. Jürgen Döllner and his team. We like to inform you about our teaching and ongoing research activities in the analysis, planning and construction of computer graphics and multimedia systems.
Matthias Trapp currently works as a post-doctoral researcher, lecturer, and head of the research group '4DnD-Geovisualization' (InnoProfile Transfer Initiative) at HPI, Computergraphics Systems Chair. His research interests include the following topics:
GPU-based image and video processing techniques and frameworks
Interactive GPU-based rendering techniques for massive spatio-temporal data
Focus+context visualization techniques for 3D virtual environments
Real-Time Animated Visualization of Massive Air-Traffic Trajectories
With increasing numbers of flights world-wide and a continuing rise in airport traffic, air-traffic management is faced with a number of challenges. These include monitoring, reporting, planning, and problem analysis of past and current air traffic, e.g., to identify hotspots, minimize delays, or to optimize sector assignments to air-traffic controllers. Interactive and dynamic 3D visualization and visual analysis of massive aircraft trajectories, i.e., analytical reasoning enabled by interactive cyber worlds, can be used to approach these challenges. To facilitate this kind of analysis, especially in the context of real-time data, interactive tools for filtering, mapping, and rendering are required. In particular, the mapping process should be configurable at run-time and support both static mappings and animations to allow users to effectively explore and realize movement dynamics. However, with growing data size and complexity, these stages of the visualization pipeline require computational efficient implementations to be capable of processing within real-time constraints. This paper presents an approach for real-time animated visualization of massive air-traffic data, that implements all stages of the visualization pipeline based on GPU techniques for efficient processing. It enables (1) interactive spatio-temporal filtering, (2) generic mapping of trajectory attributes to geometric representations and appearances, as well as (3) real-time rendering within 3D virtual environments, such as virtual 3D airport and city models. Based on this pipeline, different visualization metaphors (e.g., temporal focus+context, density maps, and overview+detail visualization) are implemented and discussed. The presented concepts and implementation can be generally used as visual analytics and data mining techniques in cyber worlds, e.g., to visualize movement data, geo-referenced networks, or other spatio-temporal data.
Colonia3D – Potenziale interaktiver Kommunikation von digitaler Denkmalpflege in öffentlichen Räumen, Prometheus Conference 2011, University of Cologne, Cologne/Germany
Kommunikation von digitaler Denkmalpflege in öffentlichen Räumen am Beispiel des Projektes Colonia3D, Invited Talk at the Department of Photogrammetry & Laserscanning at the HCU (HafenCity University), Hamburg/ Germany
High-Level Vector Editing, Invited Talk at the Computer Science Department of the Courant Institute of Mathematical Sciences at New York University, New York/USA, 2010