The objectives of this project is to be achieved based on the new technologies we
are going to develop, as well as the innovative use of existing technologies as follows:
|
I) Technologies to be developed
A human face and body detector, all necessary biometrics should be extracted by
it from surveillance video.
Besides face and gait, more visual biometric will be adopted in our system. Body
boundary and hair are considerable candidates. We will develop recognition
algorithm based on them respectively, and select the suitable biometric with
reasonable discriminability into the system.
With all these measure available, we will propose appropriate fusion strategy to
integrate them. Sometime, match score level or principal component analysis
could be adopted, and some cases, one measure could even be abandoned, for
example ignoring face measure when no face appeared.
|
II) Innovation use of existing technologies
Frontal faces will be generated by face reconstruction [2-4], which will be
adopted as input for face recognition module to get similarity score of a pair of
faces. We will select appropriate face recognition algorithm and implement it in
this project.
For gait, the gait energy image (GEI) [8], a spatio-temporal compact
representation of gait in video, will be used to characterize human-walking
properties. GEI reflects major shapes of silhouettes and their changes over gait
cycle. |
