3D Hand Biometrics

In the past decade, the interest in using 3D data for biometric person authentication has increased significantly, propelled by the availability of affordable 3D sensors. The adoption of 3D features has been especially successful in face recognition applications, leading to several commercial 3D face recognition products. In other biometric modalities such as hand recognition, several studies have shown the potential advantage of using 3D geometric information, however, no commercial­ grade systems are currently available.

The aim of this project is to fill the mentioned gap between the research and the industry. We investigate the market with the off­the­shelf low cost 3D cameras to select the best candidate based on our needs, currently Intel RealSense F200 3D camera. We demonstrate state­of­-the-­art 3D hand recognition system that combines geometric and texture information of the hand surface and is suitable for high­-end security applications as well as for small installations due to the low cost of the hardware being used.

Professor Michael Bronstein; ; PI

Jan Svoboda;

TBS Biometrics;

CTI;