Recent developments in three dimensional (3D) image acquisition systems have enabled fast and efficient generation of high quality depth images obtained from 3D surface scans, as well as, volumetric tissue images from magnetic resonance imaging (MRI) and computerized tomography (CT). In addition to surface and volumetric 3D spatial imaging, hyperspectral cameras can be considered generating 3D data with the first two dimensions in the spatial and the third dimension is in the spectral domains respectively. Recently, low cost cameras such as Kinect are providing short range 3D depth images for many applications. On the other hand, multispectral cameras are generating a rich source of hyperspectral images where each spectral image can be considered separately or all spectral images can be processed as a 3D image cube.
In this talk, latest developments in 3D image acquisition systems and their applications to Biometrics including facial expression analysis using 3D depth images will be explained. The state-of-the-art will be given in addition to our recent studies in 3D Facial Expression Recognition using local face descriptors. Our later study on Entropy based feature selection for 3D facial expression recognition will also be considered as part of the discussion.
Medical image processing is another important field for the application of 3D volumetric data acquired from MRI and CT scans, where 3D data is reconstructed using 2D slices. Many applications such as tumor detection and Alzheimer’s Disease classification are possible through the analysis of the 3D medical images. In addition to the recent methods, our latest work for the classification of Alzheimer's Disease using the MRI-based brain data from the Alzheimer's Disease Neuroimaging Initiative will be explained. Furthermore, recent applications of the hyperspectral images for Biometrics and Medicine will be explored.
Finally, possible directions of using 3D Imaging in Biometrics and Medicine will be elaborated to motivate the researchers for future research prospects.