Video Analytics from Homeland Protection to Theft Prevention and Crowd Analysis
Abstract:
This talk will provide an overview of ongoing computer vision work carried out at the
GE Global Research center. We start with a discussion of GE’s participation in the 2008
DHS STIDP trials, which was focused on the prevention of suicide bomber attacks on
large public events. A presentation of results based on a multi-camera tracking system
will be given. From a retail theft prevention perspective we will focus on deployed beta
sites, where suspicious behaviors as identified by experienced security guards have been
used to raise real time alerts. The retail environment also offers the opportunity for
auxiliary use cases, which include customer product interaction analysis. Topics such as
gaze analysis and facial expression analysis will be discussed in this context. The talk
will conclude a discussion of analytics associated with crowded sites. Topics will include
crowd segmentation at mass transit locations and the identification of social networks.
Bio:
Dr. Frederick Wheeler, General Electric, Ph.D. Rensselaer Polytechnic Institute. Dr.
Wheeler received B.S. and M.S. degrees in electrical engineering from Worcester
Polytechnic Institute, Worchester, MA, in 1990 and 1992, respectively, an M.S. degree in
Mathematics and Ph.D. degree in electrical engineering from Rensselaer Polytechnic
Institute, Troy, NY, in 1994 and 2000, respectively. At GE Global Research, he has
focused on the fields of biometrics, image super-resolution, and aerial reconnaissance.
Dr. Wheeler is the PI for the current “Multimodal Biometric Fusion” NIJ program (2007MU-CX-K001), served as PI for the completed “Active 3D Face Collection” NIJ
program (2006-92005-NY-IJ), and has led the execution of the “High Quality 3D Facial
Images from Surveillance Video” NIJ program (2005-IJ-CX-K060). He has authored
over 15 peer-reviewed publications, and has more than 15 U.S. patents pending.
Dr. Peter Tu, General Electric, Ph.D. Oxford University. Dr. Tu joined GE Global
Research in 1997. Prior to this, he was a member of the Sony Computer Vision Group in
Tokyo, Japan. He has developed a number of algorithms for latent fingerprint matching
that have been incorporated into the FBI AFIS system. Dr. Tu has also made
contributions to GE’s optical metrology systems, which are used to make high precision
3D shape measurements on manufactured parts. He has developed a number of
techniques based on Helmholtz imaging which directly addresses issues associated with
specularity and high curvature. Currently, Dr. Tu is focused on multi-view surveillance
with the aim of achieving reliable behavior recognition in complex environments.
Currently, he manages the $4 million intelligent video research effort. He has authored
more than 25 publications, has more than 20 U.S. patents pending. Dr. Tu has served as
the PI for the NIJ “High Quality 3D Facial Images from Surveillance Video,” the FBI
“ReFace” programs.