Namrata Vaswani
I completed my Ph.D. from the Department
of Electrical and Computer Engineering at the
My research interests are in estimation and detection problems in statistical signal processing, biomedical image processing and computer vision. Sometime in the future I would like to work on applied information theory problems like distributed video coding.
My research focus in the last few years has been on (a) particle filtering methods and change detection in nonlinear systems and (b) shape dynamical models and their applications in computer vision and biomedical image processing. At Georgia Tech, I am working with geometric representations of shapes of continuous curves using level set methods and their application in medical image segmentation and tracking. In the past, I have also worked on subspace methods for image classification and their error analysis and on infra-red image compression.
My Ph.D. thesis was on Change Detection in Stochastic Shape Dynamical Models with Applications in Activity Modeling and Abnormal Activity Detection. In my Ph.D., I have proposed a statistic for slow change detection in nonlinear systems and have analyzed the errors in its approximation using a particle filter. This has wide applications in slow change detection without the particle filter losing track. Also, I have proposed nonlinear systems for modeling landmark shape dynamics and have shown their application to model activity performed by a group of moving and interacting objects. Abnormal activity is detected as a slow or sudden change in the shape dynamics. Thesis Abstract
Recent Research Overview: Slides Videos Research Statement
Publications
Email: namrata AT ece DOT gatech DOT edu
Education:
Ph.D., Electrical and Computer
Engineering,
B.Tech.,
Electrical Engineering, Indian Institute of Technology (IIT),
Recent Publications: [Complete List]
Journals:
Namrata Vaswani, Rama Chellappa, Principal Component Null Space Analysis for Image and Video Classification, Accepted for IEEE Trans. on Image Processing