Institute of Computer Science — Knowledge-Based Systems Research Group
	Thomas Wiemann

	Address: University of Osnabrück Institute of Computer Science Albrechtstraße 28 Room: 31/509 49069 Osnabrück Phone: (+49) 0541 969 2438 eMail: twiemann@informatik.uni-osnabrueck.de	Fields of interest: Autonomous Mobile Robots 3D Robot/Computer Vision Sensor Data Interpretation Sensor Based Planning Artificial Intelligence
Running Projects: Las Vegas Reconstruction (Project Page) An open source software package for automatic generation of polygonal meshes from point cloud data. This software focuses on fast computation of efficient geometries suitable for 6D robot localization and is part of my PhD thesis. This software delivers tools to build surface reconstructions from point cloud data and a simple viewer to display the results. Additionally, the found surfaces will be classified into predefined categories. The main aim of this project is to deliver fast and accurate surface extraction algorithms for robotic applications such as semantic mapping, teleoperation in unknown environments and localization. The software is under permanent development and runs under Linux and MacOS. A Windows version will be made available soon. (Relatively) stable releases can be found in the download section on the project's website. Features of the current release: Surface extraction using a Marching Cubes implementation Robust normal estimation for the input data points. Surface classification. Clustering of planar patches. Export of the resulting meshes to binary .ply files that can be displayed in other software such as Meshlab. Texture generation from colored point clouds or laser scanner reflectances. Textured meshes are exported in the Wavefront .obj file format. Video export of flights through point clouds and meshes. Application Example: A point cloud taken with a Leica HDS 6000 in the forum of the "Reithalle" building. The initital Marching Cubes based surface reconstruction. Unevenness in the mesh are eliminated using a RANSAC based plane detection. A hole filling algorithm eliminated small holes resulting from sparse data and laser shadows. If the input data contains color or reflectance values, textures can be generated and rendered onto the reconstruction. The 3D Toolkit (3DTK) An open source software for automatic scan registration and fast visualization of 3D laser scans. I am currently responsible for the MacOS compatibility. TTWISS (link) I am currently an assistant in one of our external technology transfer projects. It is part of the Luftfahrtcluster Metropolregion Hamburg financed by the German Federal government as a module of their Hightech-Strategie. MERMAID (link) MERMAID (Methods for Map-Based Reasoning, Methoden zum Schlussfolgern auf Basis von Umgebungskarten) is a cooperation project between the KBS group and the MUSE research group at Massey Univ., Palmerston North, New Zealand.