Enhancing the performance of personal identity authentication systems by fusion of face verification experts

Abstract

We investigate the behavior knowledge space method (see Xu, L. et al., IEEE Transactions SMC,2, no.3, p.418-35, 1992) and decision templates method (see Kuncheva, L. et al., Pattern Recognition, vol.34, p.299-314, 2001) of classifier fusion in the context of face verification. The study involves six experts which are not only correlated, but also their performance levels differ by as much as a factor of three. Through extensive experiments on the XM2VTS database using the Lausanne protocol, we found that the behavior knowledge space fusion strategy achieved consistently better results than the decision templates method. Most importantly, it exhibited quasi monotonic behavior as the number of experts combined increased. This is a very important conclusion, as it means that the performance of the multimodal system is not degraded by adding experts

Robust face recognition using 2D and 3D data: Pose and illumination compensation

“…The fusion technique described above was shown elsewhere to give the best results (http://www.cs.colostate.edu/evalfacerec/) [12,42].…”

Face localization and authentication using color and depth images

“…The element of with the lowest value (highest similarity) corresponds to the identity of the most similar person. The fusion technique described above was shown elsewhere to give the best recognition results [30].…”

Teacher-directed learning in view-independent face recognition with mixture of experts using single-view eigenspaces

“…There are however some trainable fusion strategies such as decision templates [23] or the behavior knowledge space [24] method, proposed for person identity verification in Refs. [25,26].…”