Andreas Klaus

Segment-Based Stereo Matching Using Belief Propagation and a Self-Adapting Dissimilarity Measure

A novel stereo matching algorithm is proposed that utilizes color segmentation on the reference image and a self-adapting matching score that maximizes the number of reliable correspondences. The scene structure is modeled by a set of planar surface patches which are estimated using a new technique that is more robust to outliers. Instead of assigning a disparity value to each pixel, a disparity plane is assigned to each segment. The optimal disparity plane labeling is approximated by applying belief propagation. Experimental results using the Middlebury stereo test bed demonstrate the superior performance of the proposed method

Accurate Dense Stereo Reconstruction using Graphics Hardware

Vertex programs and pixel shaders found in modern graphics hardware are commonly used to enhance the realism of rendered scenes. Recently these hardware facilities were exploited to obtain interactive non-photorealistic effects and to perform low-level image processing tasks like texture filtering for volume visualization. We exploit modern graphics hardware to accomplish the higher level vision task of dense stereo reconstruction. In our system almost every stage of the matching procedure is executed on 3D graphics hardware, therefore utilizing the parallel vertex and pixel pipelines. Our implementation performs accurate calculations and does not suffer from the limited precision of color channels. On state of the art PC hardware our algorithm requires less than one second to reconstruct a dense mesh with subpixel accuracy for input images with one megapixel resolution.

Camera calibration from a single night sky image

We present a simple and universal camera calibration method. Instead of extensive setups we are exploiting the accurate angular positions of fixed stars. High precision is achieved by compensating the interfering error sources. Our approach uses a star catalog and requires a single input image only. No additional user input information such as focal length, exposure date or position is required. Fully automatic processing and fast convergence is achieved by performing three consecutive steps. First, a star segmentation and centroid finding algorithm extracts the sub-pixel positions of the luminaries. Second, an initial solution for the most essential parameters is determined by combinatorial analysis. Finally, the Levenberg-Marquardt algorithm is applied to solve the resulting non-linear system. Experimental results with several digital consumer cameras demonstrate high robustness and accuracy. The introduced method is advisable for applications where large calibration targets are required.

Modeling and visualizing the cultural heritage data set of Graz

The inner city (Old Town) of Graz will be the European cultural capital in 2003. In this paper we present preliminary results on the reconstruction and visualization of this kind of cultural heritage data. Starting with a simple block model obtained by converting 2 1/2 dimensional GIS (geographic information system) data we focus on the image based modeling of the facades. Herein we illustrate a robust search for corresponding points to estimate the relative orientation between image pairs.Additional, we outline our real-time rendering approach based on a LOD-R-tree concept. Special attention is paid on the LOD (level of detail) generation for historic buildings where two different ways working in 2D and 3D are explained. The visualization system is communicating asynchronously with a database system storing the LOD-R-tree data structure. This client-server configuration provides interactive navigation through the virtual scene and the scalability of a database management system.

Metropogis: a city information system

We report on a new system to augment a 3D block model of a real city obtained from aerial photogrammetry or aerial laser scanning with geo-referenced terrestrial data of the facades. The terrestrial images are acquired by a hand-held digital consumer camera. The relative orientation of the photographs is calculated automatically and fitted towards the 3D block model with minimized human input using vanishing points. The extraction of 3D primitives on the facades is based on line matching over multiple oriented images. The introduced city information system delivers a fully 3D geographic information data set and is called MetropoGIS.

VR modeler: from image sequences to 3D models

In this paper we present a novel interactive modeling system, called <i>VR Modeler,</i> to create 3D geometric models from a set of photographs. In our approach standard automatic reconstruction techniques are assisted by a human operator. The modeling system is efficient and easy to use because the user can concentrate on the 2D segmentation and interpretation of the scene whereas our system is responsible for the corresponding 3D information. Therefore we developed the user interface of VR Modeler as a monocular 3D modeling system. Additionally, we are able to obtain coarse as well as high resolution models from architectural scenes. Finally, we tested the modeling system on different types of datasets to demonstrate the usability of our approach.