The FERET Evaluation Methodology for Face-Recognition Algorithms
IEEE Transactions on Pattern Analysis and Machine Intelligence
Face Verification System Architecture Using Smart Cards
ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 1 - Volume 01
Journal of Cognitive Neuroscience
On design and optimization of face verification systems that are smart-card based
Machine Vision and Applications - Integrated Imaging and Vision Techniques for Industrial Inspection
Hi-index | 0.00 |
We discuss the effect of an optimisation strategy to be applied to image data in a smart card based face verification system. Accordingly, we propose a system architecture considering the trade-off between performance versus the improvement of memory and bandwidth management. In order to establish the system limitations, studies were performed on the XM2VTS and FERET databases demonstrating that, spatial and grey level resolution as well as JPEG compression settings for face representation can be optimised from the point of view of verification error. We show that the use of a fixed precision data type does not affect system performance very much but can speed up the verification process. Since the optimisation framework of such a system is very complicated, the search space was simplified by applying some heuristics to the problem. In the adopted suboptimal search strategy one parameter is optimised at a time. The optimisation of one stage in the sequence was carried out for the parameters of the subsequent stages. Different results were achieved on different databases, indicating that the selection of different optimum parameters for system evaluation may call for different optimum operating points.