Multi-view gender classification using symmetry of facial images

Abstract

In this paper, we propose a multi-view gender classification system with a hierarchical framework using facial images as input. The front end of the framework is a classifier, which will properly divides the input images into several groups. To ease the data sparsity problem in the multi-view scenario, facial symmetry is used to reduce the number of views. Moreover, we adopt soft assignment when dividing the input data, which can reduce the errors caused by the boundary effect in hard assignment. Then for each group, we train a gender classifier, called an expert. These experts can be any commonly used classifiers, such as support vector machines or neural networks. In this step, facial components instead of the whole face are used to achieve higher robustness against variations caused by facial alignment, illumination and occlusions. Experimental results demonstrate that our framework significantly improves the performance.

A study on different experimental configurations for age, race, and gender estimation problems

“…LBP descriptor was employed in [27] in combination with intensity and shape features to get a multi-scale fusion approach, while Ylioinas et al [28] combined it with contrast information in order to achieve a more robust classification. To extract the most discriminative LBP features, Shan [29] proposed an AdaBoost selection method and many other variants have also been proposed to get more informative features, e.g., local Gabor binary mapping pattern [30,31] and local directional pattern [32].…”

“…Finally, in [86], a study of the influence of demographic appearance in face recognition was proposed: gender (male and female), race/ethnicity (Black, White, and Hispanic), and age group (18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30), and 50-70 years old) were taken into account comparing six different solutions.…”

Assessment of deep learning for gender classification on traditional datasets

“…They conducted experiments on the LFW database, a database for studying the problem of unconstrained face recognition, which contains 13,233 color face photographs of 5,749 subjects collected from the web. A multi-view gender classification system with a hierarchical framework using facial images as input was proposed in [27] and tested on the CAS-PEAL face database. A support vector machine with automatic confidence for pattern was proposed by Zhen and Lu [29] that used a total of 10,788 different-pose facial images from the CAS-PEAL, the FERET, and the BCMI face databases.…”

A review of facial gender recognition

“…An interesting method by Manesh et al [83] uses a modified Golden Ratio template [6] to obtain the patches. In studies comparing classification rates of each individual region, the eye region has usually been found to be the most gender discriminative [77,83,134], but combining the classifiers for each region yielded better results. The effectiveness of the periocular region, which was defined as the region surrounding the eye, which may or may not include the eyebrow, and with the eyes masked out, was investigated in [79].…”