8th IEEE Workshop on Object Tracking and Classification Beyond the Visible Spectrum
CUIV Special Issue
Call for Paper
Keynote Speakers
Best Paper
Keynote Speaker

Invited Talk: Interceptors & Simulated Cool Target Tracking

Aly A. Farag
Professor of Electrical and Computer Engineering
University of Louisville

Abstract: Automatic target tracking (ATR) is a real-world problem that takes various forms on continuous basis; examples include tracking ships in the ocean, commercial airplanes, unlawful intrusions into national airspaces, and satellite systems, to name a few. Tracking and identification of targets takes another dimension with camouflage, deformation and speed, which result in lack of discriminatory features for positive identification. For example, the well-being of satellite systems may be affected by debris in space. Indeed, real world scenarios that require simultaneous identification and tracking are common and may affect the life on earth! For example, astronomers are fond of studies of comets, asteroids and other objects/phenomenon that may hit the earth and affect the human life. Perhaps space exploration provides the best examples of simultaneous identification and recognition of targets, and the use of imaging technology. Interceptors must possess intelligence to assess the type of the target, and be able to follow it for certain duration to ascertain its status before taking action. In this talk, we illustrate the simulation of cool target tracking and interception in space. Tracking is conducted with Kalman and Particle filters; recognition is based on signatures of the cool targets by IR imaging.

Bio: Aly A. Farag, PhD, received the bachelor degree from Cairo University, Egypt and the PhD degree from Purdue University in Electrical Engineering. He also holds master degrees from the Ohio State University and the University of Michigan at Ann Arbor. He joined the University of Louisville in August 1990, where he is currently a Professor of Electrical and Computer Engineering. At the University of Louisville, Dr. Farag founded the Computer Vision and Image Processing Laboratory (CVIP Lab) which focuses on imaging science, computer vision and biomedical imaging. Dr. Farag main research focus is scene analysis, object reconstruction from multimodality imaging, statistical and variational methods for object segmentation and registration. Dr. Farag has co-authored over 350 technical papers in the field of image understanding and co-edited two volumes on Deformable Models for Biomedical Applications (Springer 2007). He is the author of two upcoming textbooks: A course on Digital Signal Processing - Springer, and Statistical Models in Biomedical Image Analysis - Cambridge University Press.

Dr. Farag is an associate editor of the British Institution of Engineering and Technology Computer Vision Journal (IET-CV) and was an associate editor of the IEEE Transactions on Image Processing. He is general chair of the IEEE International Conference on Multimedia Technology (ICMT'11). He was general co-chair of the 2009 IEEE International Conference on Image Processing (ICIP-09). He is a regular reviewer for the NSF, NIH and various technical journals and international meetings. He has given number keynote presentations, tutorials, seminars and invited talks at various universities and research labs worldwide, and has been on tenure and promotion committees of several researchers in the US, Canada, Europe and the Middle East. In 2002, Dr. Farag was awarded a University Scholar designation for his technical achievements.

Invited Talk: Compressive Video Sensing

Aswin C. Sankaranarayanan
Postdoctoral research associate
Department of Electrical and Computer Engineering
Rice University

Abstract: Recent progress in compressive sensing (CS) has enabled the construction of computational imaging devices that sense parsimoniously at the information rate of the scene under view rather than its Nyquist rate. One such CS device is the single pixel camera that randomly multiplexes light spatio-temporally onto a single sensor, which enables it to operate efficiently in wavelength regimes (such as deep into the infrared) that require exotic detectors. In this talk, we report on recent work to efficiently sense and recover video from CS measurements. Our first approach relies on recovering foreground innovations from a video by performing background subtraction directly on compressive measurements. Our second approach addresses the compressive video sensing problem by regularizing the scene with a dynamical systems prior. The predictive/generative models associated with dynamical systems are employed to provide significant spatio-temporal tradeoffs at sensing. Finally, our third approach highlights some of the recent work on simultaneously sensing the scene as well as its motion as objects in the scene articulate in front of the camera. Together, these methods provide a wide range of tradeoffs on parsimony in sensing and accuracy of reconstructed signals. We demonstrate the effectiveness of our techniques on simulated and real single pixel camera data.

Bio: Aswin Sankaranarayanan, PhD, is currently a postdoctoral research associate at the ECE Department at Rice University. His research interests are broadly in the areas of computer vision and signal processing. He received his Bachelors in Electrical Engineering from the Indian Institute of Technology, Madras in 2003 and M.S and PhD. degrees from the Department of Electrical and Computer Engineering at the University of Maryland, College Park in 2007 and 2009 respectively. His thesis received the Doctoral Dissertation award from the Department of Electrical and Computer Engineering at the University of Maryland.

Copyright © OTCBVS, Contact the Webmaster.