GENDER DETERMINATION AND FACE RECOGNITION USING SIMPLE METHOD

Abstract

This system is a focused variation of a more comprehensive object recognition system. Face-related images are displayed as graphs that are tagged with topographical data and local templates. Different graphs are used to illustrate different poses.By comparing the new face with a set of computerized graphs, the "generic face knowledge," an elastic graph matching technique creates new graphs of faces. A composite image of a person's face can be created during the last phase of the matching process, and Additionally, it can identify elements that are comprised of the general face knowledge, such as gender or the existence of spectacles or a beard. A similarity function that compares the graphs makes the system effective at recognizing faces.