Anthony Cowley      Anthony Cowley
University of Pennsylvania
"Rendering Solutions to Vision Problems"
Wednesday, December 4, 4:00 PM
Packard Lab, Room 466

Abstract: Computer vision and computer graphics present a duality that suggests a tantalizing opportunity to greatly enhance robotic systems. In one direction, abstract representations of real artifacts serve as the source for computer generated imagery, while, going the other way, images captured from reality provide the inputs for computer vision. The difficulty of translating between these perspectives is felt acutely in robotic applications, where a misunderstanding involving solid geometry often leads to catastrophic failure of the robotic system. Attempts to reconcile the abstract representations robots reason with and the potentially complex scenery they encounter have traditionally revolved around constraining the inputs presented to the system by carefully engineering the environment for applications such as factory automation, or introducing long delays to allow a general purpose robot to consider its sensor inputs before acting. This hesitation before action limits the usefulness of general purpose robots, and represents a distinctive difference between traditional automation and the more adaptable robotic platforms typically found in research labs. I will describe substantial improvements to perception performance that enable new categories of tasks in the areas of navigating cluttered environments and manipulating moving objects.

Bio: Anthony Cowley is a PhD candidate in the Computer and Information Science department at the GRASP Lab at the University of Pennsylvania. He received a BSE from Penn in 2002, spent several years making video games, and was a research associate at GRASP before returning to student life in 2010. His interests are in real-time computer vision, embedded systems, multi-agent coordination, and programming language-driven verification of robotic systems.

© 2014-2016 Computer Science and Engineering, P.C. Rossin College of Engineering & Applied Science, Lehigh University, Bethlehem PA 18015.