Für eine korrekte Darstellung dieser Seite benötigen Sie einen XHTML-standardkonformen Browser, der die Darstellung von CSS-Dateien zulässt.


DAAD German Academic Exchange Service

B1 Biometrics by multispectral scattering models

Project Leader

Prof. Dr. Andreas Kolb, Prof. Dr. Volker Blanz

Initial Situation

The capturing of the external appearance of human bodies for their identification is a central issue in the context of human security and thus in the context of the Research Training Group, too. Classical imaging approaches in the field of biometrics capture finger prints, faces, the iris (and the retina), skin reflections, lip movements, ear shapes or walking movements. In almost all cases, these techniques build upon greyscale images, few use color sensors. Additionally, reflections (on surfaces) are commonly categorized as a disturbing factor and thus reduced.

Objectives and Work Plan

In contrast to prior methods, this subproject investigates the use of multi-spectral surface and subsurface scatter models for the derivation of biometrical attributes for the cases of human faces, hands and hairs. Herein, the focus lies on techniques for the identification of persons during shorter periods of time, which are robust w.r.t. cloaking additives like makeup or synthetic hair color, or which identify such conditions. The main challenges in the context of biometrics are related to the investigation of appropriate models for the concrete application scenarios and to the efficient determination of biometric parameters. The prior BRDF, BDTF and BSSRDF methods are based on a tri-stimulus approach and they are not suited for a direct derivation of multi-spectral biometric parameters.