Bio:
Dr. Horst Bischof received his M.S. and Ph.D. degree in computer science from the Vienna University of Technology in 1990 and 1993, respectively. Currently, he is Vice Rector for Research at Graz University of Technology and Professor at the Institute for Computer Graphics and Vision at the Graz University of Technology, Austria. Dr. Bischof is a member of the scientific board of Joanneum Research. He is a board member of Fraunhofer Institute IGD. His research interests include object recognition, visual learning, motion and tracking, visual surveillance and biometrics, medical computer vision, and adaptive methods for computer vision, where he has published more than 630 peer reviewed scientific papers. Horst Bischof is General Chair of CVPR 2015 and was chairman of the DAGM/OAGM conference 2012 and co-chairman of international conferences (ICANN, DAGM), and local organizer for ICPR 1996. He was program co-chair of ECCV 2006 and several times area chair of all major vision conferences. Currently, he is Associate Editor for IEEE Transactions on Pattern Analysis and Machine Intelligence, Pattern Recognition, Computer and Informatics and the Journal of Universal Computer Science. Horst Bischof is a member of the European Academy of Sciences and has received 29th Pattern Recognition award in 2002, the main price of the German Association for Pattern Recognition DAGM in 2007 and 2012, the BMVC best demo award 2012, and the best scientific paper awards at the BMCV 2007, ICPR 2008, ICPR2010, PCV 2010, AAPR2010 and ACCV 2012.
Talk topic: Visual Indoor localization for Micro Aerial vehicles
The talk will discuss methods for localizing and navigating drones in an indoor (GPS-denied) environment, eg. production site. Major emphasize will be devoted to methods using visual sensors on the micro-aerial vehicles. To illustrate the methods we will use as an example a recent project where we have used drones in production environments in a logistics context.
Bio:
Dr. Jan-Michael Frahm is an Associate Professor at University of North Carolina at Chapel Hill where he heads the 3D computer vision group. He received his Dr.- Ing. in computer vision in 2005 from the Christian-Albrechts University of Kiel, Germany. His dissertation, “Camera Self-Calibration with Known Camera Orientation” received the prize for the best Ph.D. dissertation of the year in CAU’s College of Engineering. His Diploma in Computer Science is from the University of Lübeck. His research interests include a variety of topics on the intersection of computer vision, computer graphics, robotics, and. He has over 100 peer-reviewed publications and is editor in chief for the Elsevier Journal on Image and Vision Computing.
Talk topic: Robust and Efficient Large-scale 3D Reconstruction from Crowd Sourced Imagery with Dynamic Scene Elements
To reconstruct the 3D world around us crowd sourced imagery (images and video) is the by far richest data source. There is a tremendous amount of imagery provided by photo sharing web sites, e.g. Flickr, and video sharing sites, for example YouTube, which not only covers the world’s appearance, but also reflects the temporal evolution of the world and its dynamic parts. It has long been a goal of computer vision to obtain life like virtual models from such rich imagery. The major current research challenges are the scale of the data, e.g. the recently released Yahoo 100 million-image dataset, which is still only a fraction of what is needed to fully survey the world. Further challenges are the robustness, the completeness of the registration, and the lack of data for dynamic elements. Specifically, we are currently facing significant challenges to process said imagery within a reasonable time frame given “limited” compute resources. This is particularly true as we move to personal virtual and augmented reality, which aim to explore these data at scale. The talk discusses our work on highly efficient image registration for the reconstruction of static 3D models from world-scale photo collections on a single PC in the span of six days, as well as our related work on image-based search to address the scalability. We will also discuss the efforts to overcome the challenges in achieving registration completeness and robustness. Our streaming reconstruction approach aims to ease the above challenges to achieve a 3D model from unorganized image data. Additionally, the talk will discuss our work towards overcoming the lack of data for the reconstruction of scene dynamics to achieve the goal of bringing the 3D models to life. For example, by reconstructing people and fountains, using crowd sourced imagery and videos.
Bio:
Dr. Raquel Urtasun is an Associate Professor in the department of Computer Science at the University of Toronto. She is also a Canada Research Chair in Machine Learning and Computer Vision. From January 2014 to July 2016, she was an Assistant Professor in the same department. From 2009-2014 she was an Assistant Professor at TTI-Chicago, a philanthropically endowed academic institute located in the campus of the University of Chicago. She was a visiting professor at ETH Zurich in 2010. Previously, she was a postdoctoral research scientist at UC Berkeley and ICSI and a postdoctoral associate at the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT. Raquel Urtasun completed her Ph.D. at the Computer Vision Laboratory, at EPFL, Switzerland in 2006 working with Pascal Fua and David Fleet at the University of Toronto. She has been an area chair of multiple machine learning and vision conferences (i.e., NIPS, UAI, ICML, CVPR, ECCV, and ICCV). She is on the editorial board of the International Journal of Computer Vision (IJCV), and served in the committee of numerous international conferences. Her major interests are statistical machine learning, computer vision and robotics, with a particular interest in structured prediction and their application to autonomous driving.
Bio:
Daniel Cremers received Bachelor degrees in Mathematics (1994) and Physics (1994), and a Master's degree in Theoretical Physics (1997) from the University of Heidelberg. In 2002 he obtained a PhD in Computer Science from the University of Mannheim, Germany. Subsequently he spent two years as a postdoctoral researcher at the University of California at Los Angeles (UCLA) and one year as a permanent researcher at Siemens Corporate Research in Princeton, NJ. From 2005 until 2009 he was associate professor at the University of Bonn, Germany. Since 2009 he holds the chair for Computer Vision and Pattern Recognition at the Technical University, Munich. His publications received several awards, including the 'Best Paper of the Year 2003' (Int. Pattern Recognition Society), the 'Olympus Award 2004' (German Soc. for Pattern Recognition) and the '2005 UCLA Chancellor's Award for Postdoctoral Research'. For pioneering research he received a Starting Grant (2009), a Proof of Concept Grant (2014) and a Consolidator Grant (2015) by the European Research Council. Professor Cremers has served as associate editor for several journals including the International Journal of Computer Vision, the IEEE Transactions on Pattern Analysis and Machine Intelligence and the SIAM Journal of Imaging Sciences. He has served as area chair (associate editor) for ICCV, ECCV, CVPR, ACCV, IROS, etc, and as program chair for ACCV 2014. He serves as general chair for the European Conference on Computer Vision 2018 in Munich. In December 2010 he was listed among “Germany's top 40 researchers below 40” (Capital). On March 1st 2016, Prof. Cremers received the Leibniz Award 2016, the biggest award in German academia. He is Managing Director of the Department of Computer Science.
Talk topic: Dense & Direct Methods for 3D Reconstruction & Visual SLAM
The reconstruction of the 3D world from images is among the central challenges in computer vision. Starting in the 2000s, researchers have pioneered algorithms which can reconstruct camera motion and sparse feature-points in real-time. In my talk, I
will introduce spatially dense methods for camera tracking and 3D reconstruction which do not require feature point estimation, which exploit all available input data and which recover dense or semi-dense geometry rather than sparse point clouds. Applications include 3D photography, 3D television, and autonomous vehicles.
Bio:
Davide Scaramuzza is Assistant Professor of Robotics at the University of Zurich. He is founder and director of the Robotics and Perception Group, where he does cutting-edge research on low-latency computer vision applied to the autonomous navigation of visually-guided ground and micro flying robots. He received his PhD (2008) in Robotics and Computer Vision at ETH Zurich. He was Postdoc at both ETH Zurich and the University of Pennsylvania. From 2009 to 2012, he led the European project SFLY (Swarm of Micro Flying Robots). He was awarded the IEEE Robotics and Automation Early Career Award (2014), the ERC Starting Grant (2014, through the SNSF), a Google Research Award (2014), the European Young Researcher Award (2012), and the Robotdalen Scientific Award (2009). He is coauthor of the 2nd edition of the book "Introduction to Autonomous Mobile Robots" (MIT Press). He is author of the first open-source Omnidirectional Camera Calibration Toolbox for MATLAB, which, besides accomplishing thousands of downloads worldwide, is also used at NASA, Philips, Bosch, and Daimler. He is also author of the 1-point RANSAC algorithm, an effective and computationally efficient reduction of the standard 5-point RANSAC for visual odometry, when vehicle motion is non-holonomic. He is an Associate Editor for the IEEE Transactions of Robotics. His research interests are field and service robotics, intelligent vehicles, and computer vision. Specifically, he investigates the use of cameras as the main sensors for robot navigation, mapping, exploration, reasoning, and interpretation. His interests encompass both ground and flying vehicles.
Talk topic: Event based Vision for High Speed and Low-Power Localization and Mapping
Event-based cameras are revolutionary vision sensors with four key advantages: a measurement rate that is almost 1 million times faster than standard cameras, a latency of microseconds, a high dynamic range that is eight orders of magnitude larger than that of standard cameras, and power consumption that is 100 less than that of a standard camera. Event-based sensors open frontiers which are unthinkable with standard cameras (which have been the main sensing technology of the past 50 years). These revolutionary sensors enable the design of a new class of algorithms to track a baseball in the moonlight, build a flying robot with the same agility of a fly, localizing and mapping in challenging lighting conditions and at remarkable speeds. These sensors became commercially available in 2008 and are slowly being adopted in mobile robotics and computer vision. They covered the main news in 2016 with Intel and Bosch announcing a $15 million investment in event-camera company Chronocam and Samsung announcing its use with the IBM's brain-inspired TrueNorth processor to recognize human gestures. This talk will cover the sensing hardware as well as the algorithmic methods needed to take advantage of these sensors.
