||Dr. Anuj Srivastava|
Department of Statistics
Florida State University
Statistical Shape Analysis and Modeling Group (SSAMG)
|Office: 106D OSB, FSU, Tallahassee, FL|
Phone: (850) 644-8832
Fax: (850) 644-5271
Email: anuj at stat.fsu.edu
My Google Scholar Profile can be found
here , and DBLP listing
My Plenary Talk at ICIP 2013 in Melbourne, Australia (without movies)
Talk (PDF File)
I am a professor in the Department of Statistics, Florida State University.
Here is my brief intro. I obtained M.S. and Ph.D. degrees in electrical engineering from Washington University in St. Louis, in the years 1993 and 1996, respectively, both under the guidance of
Prof. Michael I. Miller (now at the Johns Hopkins University). During 1996-97, I was a visiting research scientist at the Division of Applied Mathematics, Brown University. In Fall 1997, I joined the Department of Statistics at the Florida State University as an assistant professor. During 2003-2006, I was an associate professor, and starting Fall 2007 I am a professor here at FSU. I was awarded the Developing
(see a description here) in 2005 and the Graduate Mentor Faulty Award in 2008.
During my graduate studies and postdoctoral stay at Brown University, I got a chance to work closely with and learn from
Prof. Ulf Grenander . Over the last three decades, his development of metric pattern-theory has been both profound and powerful. An important aspect of this approach is the broad range of the knowledge base that it uses--
algebra, geometry, statistics, computational science, and imaging science.
Grenanderís pattern theory has been a major influence on my research and approach.
My main interest lies in the area of statistical image understanding with a focus on fundamental issues.
Here one seeks to design computerized systems for understanding observed scenes through camera images,
much like our own human vision system. This technology is needed in many applications --
medical diagnosis, biometrics, video surveillance, undersea imaging, terrain mapping, and satellite
image analysis. Imaging devices have become ubiquitous in today's society, and image data has
become one of the most common sources of information. Consequently, statistics, the traditional science of
data analysis, has an important role to play in developing automated systems that can understand
images and their contents.
Our approach is to develop representations for objects of interest, by studying their
shapes, textures, appearances, and motions. Using probability models on these representations, learned from
past data, we use Bayesian strategies for deriving inferences from given image data.