•   When: Monday, November 07, 2016 from 10:00 AM to 11:00 AM
•   Speakers: Jana Kosecka, Department of Computer Science, George Mason University
•   Location: JC Room B
•   Export to iCal

Abstract

Advancements in robotic navigation, mapping, object search and recognition rest to a large extent on robust, efficient and scalable understanding of the surrounding environment.  I will discuss several approaches we have developed for capturing geometry and semantics of environment from video, RGB-D data, or just simply a single RGB image, focusing on indoors and outdoors environments relevant for robotics applications. I will demonstrate our work on predicting locations of generic objects in videos acquired by a moving vehicle, localization, and more recent results using deep convolutional neural networks (CNNs) for detailedsemantic parsing, 3D reconstruction and object detection from single RGB image. The
applicability of the presented techniques for autonomous driving, service robotics, mapping and augmented reality applications will be highlighted.

Speakers Bio

Jana Kosecka is an Associate Professor at the Department of Computer Science, George Mason University. She obtained her M.S.E. in Electrical Engineering and Computer Science from Slovak Technical University and Ph.D. in Computer Science from University of Pennsylvania in 1996. In 1996 - 1999 she was a postdoctoral fellow at the EECS Department at University of California, Berkeley. She is the recipient of David Marr's prize (with Y. Ma, S. Soatto and S. Sastry) and received the National Science Foundation CAREER Award. Jana is an Associate Editor of IEEE Transactions on Robotics and a Member of the Editorial Board of International Journal of Computer Vision. Her general research interests are in Robotics and Computer Vision. In particular she is interested 'seeing' systems engaged in autonomous tasks, acquisition of static and dynamic models of environments by means of visual sensing and human-computer interaction.