Here, the occluded regions in the p

Here, the occluded regions in the probe faces are restored according to the gallery ones. For instance, Bagchi et al. [4] proposed to restore facial occlusions. The detection of the occluded regions is carried out by thresholding the depth map values of the 3D image. Then the restoration is taken on by Principal Component Analysis (PCA) [31]. There are also several approaches that rely on the estimation of the occluded parts. Drira et al. [6] applied a statistical shape model to predict and restore the partial facial curves. Iterative closest point (ICP) algorithm has been used to remove occluded regions in [8]. The restoration is applied using a curve, which uses a statistical es- timation of the curves to manage the occluded parts. Partially observed curves are completed by using the curves model produced through the PCA technique.