Age invariant face recognition using convolutional neural networks and set distances

An age invariant face recognition method comprising the steps of:

• acquiring by an input device a representation of a facial image;
• receiving said representation of the facial image by a pre-trained convolutional neural network and extracting compact, highly discriminative and interoperable feature descriptors;
• building a biometric gallery during an enrollment process when biometric traits of all known subjects are extracted as feature descriptors and stored as sets of image templates in a database;
• receiving by a classifier said feature descriptors from the convolutional neural network and comparing said feature descriptors with feature descriptors of subjects enrolled in the gallery, choosing set similarity distances by the classifier from the group consisting of minimum distance, directed Hausdorff distance, undirected Hausdorff distance, directed modified Hausdorff distance, and undirected modified Hausdorff distance;
• and
• generating a match or no match by an output device of face recognition subject to aging, a match being used for authentication or identification.