Paper Title
Design And Development Of Rotation And Scale Invariant 3d Models For Efficient Representation And Accurate Classification Of Human Faces

Face recognition, an important biometrics technique, plays a critical role in many real-world multimedia applications. Despite being studied extensively in literature, existing face recognition techniques still suffer from a lot of challenges when being applied in real-world applications. In particular, many 2D face recognition approaches work excellently under well-controlled conditions, but their recognition accuracy often decrease considerably when handling realworld face recognition tasks where variations are common for pose, illumination and expression.On the other hand, along with the advances of various 3D capture devices, 3D face recognition techniques are receiving more and more research attention. The highly detailed 3D mesh data can capture rich information, which potentially provides much more clues to tackle some unsolved challenges in 2D face recognition tasks, especially for pose and illumination variations.Following this direction, in this paper, we investigate a number of 3D face recognition scheme that addresses the open challenge of Pose Invariant Face Recognition. In particular, we survey an effective approach to tackling the pose invariant 3D face recognition task, which is equipped with a set of effective 3D parametrization, alignment, and spares feature representation techniques. Index Terms— Face recognition, expression recognition, face de-identification, facial expression, pose-invariant recognition, 3-D face classification.