System and method for identifying faces in unconstrained media

摘要

A method and system for facial recognition is provided. The method includes determining a three-dimensional (3D) model of the individual's face based on different images of the individual. The method also includes extracting a two-dimensional (2D) patch from the 3D model. Further, the method includes generating a plurality of facial signatures using different combinations of 2D patches, the plurality of signatures corresponding to respective views of the 3D model from different angles.