Invited speakers (in alphabetical order):
Bio: Herbert Bay holds a PhD in Computer Vision Studies from the ETH Zurich (2006) where he developed the famous SURF algorithm for image feature detection and description used for image recognition, augmented reality as well as various other computer vision applications. Before his PhD, he received a Master's Degree in Microengineering from EPF Lausanne. After his graduation, he joined IBM Switzerland where he was thoroughly trained in project management, sales, and communication. Soon he was helping customers address their critical business challenges for e-learning projects as an Engagement Manager. He was also entrusted with teaching internal communication courses in Zurich and Geneva. During his time at kooaba, Herbert followed an entrepreneurial workshop at Babson College, in Boston MA (2007) as well as an Executive MBA workshop in Palo Alto, CA (2008) held by the renown business school IMD. Besides his passion for Internet startups and entrepreneurship, Herbert is fund of offshore sailing.
Talk: In this keynote, some near-future commercial opportunities where computer vision will have a fundamental impact are presented. Besides Augmented Reality, numerous other possibilities exist on how computer vision can change business models and customer behavior. We will discuss opportunities and challenges for some concrete examples such as image-based online ads, automatic photo management apps, etc.
Norbert Brändle
AIT Austrian Institute of Technology
Title: Vision Impossible? A report about coupling pedestrian traffic modeling with real world data from visual surveillance
Bio: Norbert Brändle graduated in Computer Sciences and received his PhD in computer vision from the Vienna University of Technology, Austria. Norbert worked as a researcher at the Novartis Research Institute Vienna and was a faculty member at the Vienna University of Technology, Pattern Recognition and Image Processing Group. Since 2004 Norbert has been part of the Dynamic Transportation Systems team at the AIT Mobility department. As a senior researcher and deputy head of the business unit, Norbert currently coordinates research projects in the areas of multimodal traffic data acquisition and analysis.
Talk: Pedestrian traffic modeling can provide valuable insights into construction projects such as shopping centers, airports, railway stations, or stadiums. In order to calibrate and validate such models, real world data about pedestrian traffic is indispensable, and visual surveillance has the potential to provide the most detailed data compared to other technologies. This talk reports about efforts of a transportation research group to apply state-of-the art methods for automated human motion analysis, and the ways to deal with their limits.