Overview

The system uses a 3D scanner which acquires a 3D scan and a 2D photograph, these two images are then superimposed. The system automatically cleans the data and corrects the image (for e.g. in the presence of pose variations). The 2D and 3D features are extracted and fused.

In the case of verification/authentication, the system applies a recognition algorithm to match the 3D scan of the person to a previously stored scan of the same person (one-to-one matching). In the case of identification, the features extracted from the 3D scan of the person are matched with the features of all the other 3D scans in the database (one-to-many matching).