Enhanced constrained local models for gender prediction

Abstract

Face land-marking, defined as the detection and positioning of distinctive characteristics, is a crucial goal shared by various organizations, ranging from biometric recognition to mental state comprehension. Despite its apparent simplicity, this problem has been extensively investigated because of inherent face variability and a variety of confusing variables such as posture, voice, illumination, and occlusions. In this paper, an integrated mount model is created to increase the power of constrained local models, and a ground-breaking result for feature detection is obtained using this model. Furthermore, four classifiers have been used in the level of gender prediction. The results of the experiment showed that the proposed model performs admirably.