Item

Abstract

In this paper we propose a computational architecture of face recognition based on evidence from cognitive research. Using an implementation of this architecture we were able to model aspects of human performance, which were found in psychophysical studies. Furthermore, results from additional recognition experi ments show that our framework is able to achieve excellent recognition performance even under large view rotations. Thus, our study is an example of how results from cognitive research can be used to construct recognition systems with better performance. Finally, our results also make new experimental predictions, which can be tested in further psychophysical studies thus closing the loop between experiment and modeling.