Compressive sensing is a new technology that allows one to acquire data with high information content, based on measurements that are much fewer in number than would be expected by Nyquist theory. This is achieved by leveraging signal models that go beyond signal bandwidth, to exploit the fact that signals of interest typically reside in a low-dimensional subspace. This includes exploitation of concepts like sparsity, unions of subspaces, and low-rank signal models, as well as nonlinear signal-recovery methods. In this talk we examine how modern statistical tools may be used to achieve remarkable levels of compression directly at the measurement point, with demonstrations based on new hyperspectral and video cameras. The talk will examine the fundamental mathematics and statistics, and show results based on real cameras we have developed and tested.
Lawrence Carin earned the BS, MS, and PhD degrees in electrical engineering at the University of Maryland, College Park, in 1985, 1986, and 1989, respectively. In 1989 he joined the Electrical Engineering Department at Polytechnic University (Brooklyn) as an Assistant Professor, and became an Associate Professor there in 1994. In September 1995 he joined the Electrical Engineering Department at Duke University, where he is now the William H. Younger Distinguished Professor. Dr Carin's early research was in the area of electromagnetics and sensing, and over the last ten years his research has moved to applied statistics and machine learning. He has recently served on the Program Committee for the following machine learning conferences: International Conf. on Machine Learning (ICML), Neural and Information Processing Systems (NIPS), Artificial Intelligence and Statistics (AISTATS), and Uncertainty in Artificial Intelligence (UAI). He was previously an Associate Editor of the IEEE Trans. on Antennas and Propagation, and he is currently an Associate Editor for the IEEE Trans. on Signal Processing and the SIAM J. of Imaging Science. He is an IEEE Fellow.
Researchers should cite this work as follows: