April 21, 2017, Ulugbek Kamilov, Mitsubishi Electric Research Laboratory
Friday, April 21, 2017, 3pm-4pm
8 St. Mary’s Street, PHO 211
Refreshments at 2:45pm
Mitsubishi Electric Research Laboratory
SEAGLE: Robust Computational Imaging under Light Scattering
Current methods in high-resolution three-dimensional (3D) optical microscopy rely on linear scattering models that assume weakly scattering samples, making them inherently inaccurate for many applications. This places fundamental limits—in terms of resolution, penetration, and quality—on the imaging systems relying on such models. In this talk, we describe a new technique for computational imaging called SEAGLE that combines a nonlinear scattering model and a regularized inversion algorithm. SEAGLE exploits an efficient representation of light scattering as a recursive neural network for formulating a fast, large-scale imaging algorithm. The key benefit of SEAGLE is its efficiency and stability, even for objects with large permittivity contrasts. SEAGLE is suitable for robust imaging under multiple light scattering and has a potential to broadly impact 3D imaging of multicellular organisms such as biological tissue.
Ulugbek S. Kamilov is a Research Scientist in the Computational Sensing team at Mitsubishi Electric Research Laboratories (MERL), Cambridge, MA, USA. Dr. Kamilov obtained his B.Sc. and M.Sc. in Communication Systems, and Ph.D. in Electrical Engineering from the École Polytechnique Fédérale de Lausanne (EPFL), Switzerland, in 2008, 2011, and 2015, respectively. In 2007, he was an Exchange Student at Carnegie Mellon University (CMU), Pittsburgh, PA, USA, in 2010, a Visiting Student at Massachusetts Institute of Technology (MIT), Cambridge, MA, USA, and in 2013, a Visiting Student Researcher at Stanford University, Stanford, CA, USA. Since 2016, Dr. Kamilov is a member IEEE Special Interest Group on Computational Imaging. Dr. Kamilov’s is interested in enabling the imaging and analysis of previously inaccessible information by designing computational techniques that can be scaled to large data volumes. In particular, he is interested in imaging through scattered media, multimodal sensing, convex and non-convex optimization, traditional and convolutional dictionary learning, Bayesian estimation, and imaging inverse problems in most forms and disguises. He has co-authored 17 journal and 32 conference publications in these areas. His thesis work on Learning Tomography (LT) was selected as a EPFL Doctorate Awards 2016 finalist and was featured in Nature News and Views.
Faculty Host: Vivek Goyal
Student Host: Yue Zhang