| Anshuman Razdan
Professor, Director of Advanced Technology Innovation Collaboratory
Arizona State University, Arizona
Biography:
Dr. Anshuman Razdan is Associate Professor in the Division of Computing Studies and the Director of Advanced Technology Innovation Collaboratory (ATIC,) and the I3DEA Lab (i3deas.asu.edu) at Arizona State University. He has been a pioneer in computing based interdisciplinary collaboration and research at ASU. His research interests include Geometric Design, Computer Graphics, Document Exploitation and Geospatial visualization and analysis. He is the Principal Investigator and a collaborator on several federal grants from agencies including NSF, NGA and NIH. He has a BS and MS in Mechanical Engineering and Ph D in Computer Science.
Talk Title: 3D Shape Searches for Biometrics and Other Practical applications
Abstract: The talk will focus on two topics related to 2D/3D shape and searching. The first part will focus on an application where we are using (curve and surface) feature extraction techniques. The application is 3D face authentication and recognition and some recent results will be shared. Our feature extraction method can handle data with different facial expressions, is robust to occlusions caused by hair on the head or face, and successful when upper body, clothing, etc. are included in the scan. Pose and orientation need not be known apriori. Development of weighted features for discriminant analysis will also be discussed.
The second part of the talk will deal with the widely applicable problem of 2D curve matching. Our recent research in this area and experimental results from our work will be presented. Specifically, our technique based on a special invariant signature that can handle the problem of matching open curves. The method is scale invariant and can do whole-to-part and part-to-part matching. The former in and the latter in time with space requirement.
|