Instructors: Christopher Geyer (University of California, Berkeley), Tomas Pajdla (Czech Technical University, Prague), and Kostas Daniilidis (University of Pennsylvania).
Duration: 3.5 hours
In robotics it is essential for many tasks to have complete awareness of the environment. For example, in order for aerial or ground robots to avoid obstacles, evade enemies, or generally to perceive and plan in an environment, it is very useful to have vision sensors capable of seeing in all directions. In this short course we present an introduction to the growing field of omnidirectional vision. Below is an outline of some of the topics we plan to cover:
Christopher Geyer is currently a post-doc at the University of California, Berkeley. He received his PhD from the University of Pennsylvania under the supervision of Kostas Daniilidis, and received the Rubinoff award for an outstanding dissertation that has resulted in or could lead to innovative applications of computer technology. His interests are in applications of omnidirectional cameras to scene reconstruction and the control of autonomous aerial and ground vehicles. He is also the winner of the A. Atwater Kent Prize in Electrical Engineering from the University of Pennsylvania.
Tomas Pajdla (MSc, PhD from the Czech Technical University in Prague) is an Assistant Professor at the Czech Technical University in Prague, and a lecturer of computer vision and robotics. In 1994-1995, prior to his current appointment, he worked with Luc Van Gool at Katholieke Universiteit Leuven. T. Pajdla works on panoramic imaging since 1996. He coauthored works that introduced epipolar geometry of panoramic cameras, investigated the use of panoramic images for robot localization, contributed to studies of panoramic mosaics, and he proposed to study omnidirectional cameras that do not have a center of projection but have a generalized epipolar geometry. He is a Member of IEEE, ACM, and Czech Pattern Recognition Society. He was awarded the best paper Prize at OAGM'1998. In 2002 he received the award for the best scientific paper at BMVC'2002. He serves as the programme committee chair of ECCV 2004.
Kostas Daniilidis is Associate Professor of Computer and Information Science at the University of Pennsylvania where he was Assistant Professor from 1998 to 2003. He is a member of the interdisciplinary GRASP laboratory. He obtained his MSE (Diploma) in Electrical Engineering from the National Technical University of Athens, 1986, and his PhD (Dr.rer.nat.) in Computer Science from the University of Karlsruhe, 1992, under the supervision of Hans-Hellmut Nagel. Prior to his current appointment, he has been with the University of Kiel. His research interests are in space and motion perception with machines, with focus on omnidirectional vision and new camera technologies. He was the chair of the 2000 IEEE Workshop on Omnidirectional Vision and co-editor of the Special Issue of the IEEE Robotics and Automation Magazine on Panoramic Robotics. He is Associate Editor of the IEEE Trans. Pattern Analysis and Machine Intelligence. He was the 2001 recipient of the Ford Motor Company Award for Best Faculty Advising at the University of Pennsylvania.