Face Recognition Based on 2D and 3D Features

This paper presents a completly automated face recognition system integrating both two dimensional (texture) and three dimensional (shape) features. We introduce a novel fusion strategy that allows to automatically select, for each face, the most relevant features from each modality. The performance is evaluated on the largest public data corpus for face recognition currently available, the Face Recognition Grand Challenge version 2.0.

• Type: Conference
• Year: 2007
• Where: KES
• Authors: Stefano Arca, Raffaella Lanzarotti, Giuseppe Lipori
• Keywords: Automated Face Recognition, Face Recognition, KES 2007, Recognition Grand Challenge