Research

A webpage with introductory material about astronomy in general and references to our projects related to the telescope housed in our observatory, on the roof of the institute. Latest photographs and news about public star parties will be posted on this page, also.

We present a video processing algorithm for texture replacement of moving garments in monocular video recordings as an extension of our multi-view capture method. A time-coherent texture interpolation is obtained by the use of 3D radial basis functions. Shading maps are determined with a surface reconstruction technique and applied to new textures which replace the color pattern in the video sequence.

We present a physically based approach to compute the colors of the sky during the twilight period before sunrise and after sunset. The simulation is based on the theory of light scattering by small particles. A realistic atmosphere model is assumed, consisting of air molecules, aerosols, and water. Air density, aerosols, and relative humidity vary with altitude. In addition, the aerosol component varies in composition and particle size distribution. This allows us to realistically simulate twilight phenomena for a wide range of different climate conditions. Besides considering multiple Rayleigh and Mie scattering, we take into account wavelength-dependent refraction of direct sunlight as well as the shadow of the Earth.

We propose a system which projects images of astronomical objects (with focus on nebulae and galaxies), animations or additional information directly into the eyepiece view of an astronomical telescope. As the telescope orientation is tracked continuously, the projected image is adapted in real-time to the object which is currently visible through the eyepiece. This way, visitors to public observatories have the possibility to experience the richness of deep sky objects while directly gazing at them through a telescope.

We present a new image-based algorithm for surface reconstruction of moving garment from multiple calibrated video cameras. Using a color-coded cloth texture, we reliably match circular features between different camera views. The deforming geometry can be used for different graphics applications, e.g. for realistic retexturing.

We have developed an interactive visualization tool for rendering physically correct representations of arbitrary dust distributions surrounding a central illuminating star. Our algorithm can be used to create virtual fly-throughs of a reflection nebula for either interactive desktop visualizations or scientifically accurate animations for planetarium shows. Besides being able to take the measured data from a given reflection nebula as input and producing a plausible simulation of the nebula's appearance from arbitrary viewpoints, the system also supports the investigation of the visual effects of changing a nebula's physical parameters.

Athletes and coaches in most professional sports make use of high-tech equipment to analyze and, subsequently, improve the athlete's performance. High-speed video cameras are employed, for instance, to record the swing of a golf club or a tennis racket, the movement of the feet while running, and the body motion in apparatus gymnastics. High-tech and high-speed equipment, however, usually implies high-cost as well. In this paper, we present a passive optical approach to capture high-speed motion using multi-exposure images obtained with low-cost commodity still cameras and a stroboscope. The recorded motion remains completely undisturbed by the motion capture process.

Determining the three-dimensional structure of distant astronomical objects is a challenging task, given that terrestrial observations provide only one viewpoint. For this task, bipolar planetary nebulae are interesting objects of study because of their pronounced axial symmetry (due to the physics of the interacting gas flows), and because the glowing photo-ionized gas comprising the nebula exhibits negligible absorption. Making use of these properties, we present a technique to automatically recover the three-dimensional structure of bipolar planetary nebulae from conventional two-dimensional images.

In traditional video, the viewpoint in a scene is the one chosen by the director. It cannot be changed by the viewer while he is watching the video. The goal in free-viewpoint video, on the other hand, is to create a feeling of immersion by offering the viewer the opportunity to interactively change its viewpoint in the scene. The human body and its motion plays a central role in most visual media and its structure can be exploited for robust motion estimation and efficient visualization. In our work we develop a system that uses multi-view synchronized video footage of an actor's performance to estimate motion parameters and to interactively re-render the actor's appearance from any viewpoint.

Non-invasively determining the three-dimensional structure of real flames is a challenging task. We present a tomographic method for reconstructing a volumetric model from multiple images of fire. The method is similar to sparse-view computerized tomography and is applicable to static camera setups observing dynamic flames. Using an algebraic reconstruction method, we can restrict the solution space such that a high quality model is obtained from only a small number of camera images. An additional advantage is fast processing of multi-video sequences to generate time-varying models for animation purposes.

This work concentrates on rendering the solar disc considering Rayleigh scattering, Mie scattering, absorption, and refraction. The atmosphere is modeled in layers, each layer having a set of individual optical properties. Based on different atmospheric temperature profiles and climates, the solar disc is rendered in realistic shape and color. In particular, we replicate optical phenomena such as the red and the green flash, limb darkening, and refractive distortions of the solar disc.

We model the dynamic geometry of a time-varying scene as a 3D isosurface in space-time. The intersection of the isosurface with planes of constant time yields the geometry at a single time instant. An optimal fit of our model to multiple video sequences is defined as the minimum of an energy functional. This functional is given by an integral over the entire hypersurface, which is designed to optimize photo-consistency. A PDE-based evolution derived from the Euler-Lagrange equation maximizes consistency with all of the given video data simultaneously. The result is a 3D model of the scene which varies smoothly over time.

We have developed an algorithm for capturing the motion of deformable surfaces, in particular textured cloth. In a calibrated multi-view camera setup, we determine the optical flow between consecutive video frames and derive 3D vertex motion from optical flow. The resulting velocity field can cause distortions of the triangulated surface. We use a deformable surface model with constraints for vertex distances and curvature for smoothing and regularisation. Tracking errors in long video sequences are corrected by a silhouette matching procedure.