Weekly Meeting - Dr. Octavia Camps, "Teaching Computers to See"
Click title for more info
When: February 12th, 2014 at 11:45am
Where: 308 Snell Engineering
This week we'll have Professor Camps coming in to talk to us about her research in computer vision. As usual there will be pizza and drinks, see you then!
We live in a world overflowing with data collection and non-stop communications. Cameras are ubiquitous everywhere and hold the promise of significantly changing the way we live and interact with our environment. Dynamic vision systems are uniquely positioned to address the needs of a growing segment of the population. Smart environments that are aware of user activities can enable an aging population to carry on independent lives for as long as possible. Computers that interpret facial expressions to obtain cues to user confusion can lead to simpler interfaces. Finally, activity-monitoring systems capable of recognizing and correlating actions at different locations can improve security and reduce the time response to emergencies. However, a major roadblock in taking full advantage of the exponential growth in data collection and actuation capabilities stems from the curse of dimensionality. As an example, a short video sequence from a single camera contains mega bytes of (highly redundant) data. Similar situations arise when dealing with time-traces of gene promoters in systems biology. Simply put, existing techniques are ill equipped to deal with the resulting overwhelming volume of data.
This talk discusses the key role that systems theory can play in timely extracting and exploiting actionable information that is very sparsely encoded in high dimensional data streams. The central theme of this approach is the use of dynamical models as information encoding paradigms. Our basic premise is that spatio/temporal dynamic information can be compactly encapsulated in dynamic models, whose rank, a measure of the dimension of useful information, is often far lower than the raw data dimension. Embedding problems in the conceptual world of dynamical systems makes available a rich, extremely powerful resource base, leading to robust solutions, or, in cases where the underlying problem is intrinsically hard, to computationally tractable approximations with suboptimality certificates. These ideas are illustrated with several applications, including contraflow detection for airport security, multi- target tracking and activity recognition.
Bio: Octavia Camps received a B.S. degree in computer science and a B.S. degree in electrical engineering from the Universidad de la Republica (Uruguay), and a M.S. and a Ph.D. degree in electrical engineering from the University of Washington. Since 2006, she is a Professor in the Electrical and Computer Engineering Department at Northeastern University. From 1991 to 2006 she was a faculty of Electrical Engineering and of Computer Science and Engineering at The Pennsylvania State University. In 2000, she was a visiting faculty at the California Institute of Technology and at the University of Southern California and in 2013 she was a visiting faculty at Boston University. Her main research interests include robust computer vision, image processing, and machine learning. She is a former associate editor of Pattern Recognition and Machine Vision Applications. She is a member of the IEEE society