Keynote speaker

#
Peter Wonka
Bio

Dr. Peter Wonka is Associate Director of the Visual Computing center (VCC) at King Abdullah University of Science and Technology (KAUST) and Professor in the Computer Science program. Dr. Wonka received his doctorate from the Technical University of Vienna in computer science. Additionally, he received a Masters in Urban Planning and a Masters in Computer Science from the same institution. After his PhD, Dr. Wonka worked as postdoctoral researcher at the Georgia Institute of Technology and as faculty at Arizona State University. His research interests include various topics in computer graphics, visualization, remote sensing, computer vision, image processing, machine learning, and data mining.

Abstract Encoding Prior Knowledge for Urban Reconstruction

In this talk different strategies to encode prior knowledge about building shapes for urban reconstruction are presented. After initial stages in the reconstruction pipeline, e.g. using laser scanning or visible images, we are often confronted with noisy and incomplete data. In order to clean up and complete the data, one possible strategy is to make strong assumptions about the shape of building mass models and facade layouts. The goal of this type of reconstruction is to create plausible high quality models. First, shape grammars that are used in procedural modeling to create large city models are discussed and the initial CGA-shape grammar including multiple extensions is presented along with different attempts to use grammars to help in urban reconstruction. The talk also shows the utility of graphical models to encode information about facade layouts and building mass models and applications to missing data completion.