We, in collaboration with the CMRI Electro-Optics Technology Group, are developing and evaluating human identification technologies as part of the Defense Advanced Research Projects Agency (DARPA) sponsored program in Human Identification at a Distance (HumanID). Our efforts can be roughly divided into the following categories:
Static and Dynamic Body Biometrics
We are developing algorithms for the use of 3D space-time body biometrics for human identification. Our research focuses on two main issues: (1) the extraction of body biometric features from video and (2) identifying subjects from the extracted body biometric features.
Face Recognition From Video
We are developing face recognition algorithms that take full advantage of the additional information contained in video imagery.
Human Identification From Expression
We are investigating the extent to which facial expression impacts upon existing face recognition algorithms, and also the extent to which facial expressions and other spatio-temporal information can be used to identfy people.
We are invesitgating the information content of hyperspectral and polarimetric imagery for human identification.
We are collecting a number of databases to aid in the evaluation of human identification algorithms. In particular, we are collecting databases of: (1) synchronized multi-camera video sequences of body motion, (2) human faces under a wide range of imaging conditions, (3) AU coded expression videos, and (4) hyperspectal and polarimetric images of human faces.
|The Robotics Institute is part of the School of Computer Science, Carnegie Mellon University.|
Contact Us | Update Instructions