High level imaging for face recognition towards more accurate detection
Narasimhan Renga Raajan, Mohankumar Vishnu Priya, S. Suganya, Durairaj Parthiban, A. Jenifer Philomina, M. S. Ramkumar, B. Monisha · IEEE-International Conference On Advances In Engineering, Science And Management · 2012
Three-dimensional face recognition is an emerging technique in which 3D geometry of the face is used for accurate identification of the faces. The geometry of the face changes significantly as the outcome of facial expressions yielding in good efficiency of the technique. Using stereoscopy, the depth map in an image is determined and the recognition of the face is performed. For this, we introduce a technique called AAM (Active Appearance Model) which is a powerful statistical model for modeling where the face is identified based on their feature points even though if there are many external factors to which they are very sensitive. It is very accurate in alignment and effective for handling face deformation. Then the face is projected in 3D for higher efficacy. This technique can be implemented in real time applications such as digital cameras, web cameras and mobile phones.