Automatic age and gender classification using supervised appearance model

Abstract

YesAge and gender classification are two important problems that recently gained popularity in the\ud research community, due to their wide range of applications. Research has shown that both age and gender\ud information are encoded in the face shape and texture, hence the active appearance model (AAM), a statistical\ud model that captures shape and texture variations, has been one of the most widely used feature extraction\ud techniques for the aforementioned problems. However, AAM suffers from some drawbacks, especially when\ud used for classification. This is primarily because principal component analysis (PCA), which is at the core of\ud the model, works in an unsupervised manner, i.e., PCA dimensionality reduction does not take into account\ud how the predictor variables relate to the response (class labels). Rather, it explores only the underlying structure\ud of the predictor variables, thus, it is no surprise if PCA discards valuable parts of the data that represent discriminatory\ud features. Toward this end, we propose a supervised appearance model (sAM) that improves on AAM\ud by replacing PCA with partial least-squares regression. This feature extraction technique is then used for the\ud problems of age and gender classification. Our experiments show that sAM has better predictive power than the\ud conventional AAM

Computational Intelligence in Automatic Face Age Estimation: A Survey

“…These include Locality Sensitive Discriminant Analysis (LSDA), Marginal Fisher Analysis (MFA) [39], PCA, Neighborhood Preserving Projections (NPP), Locality Preserving Projections (LPP), and Orthogonal LPP (OLPP) [31], [46]. However, Active Appearance Model (AAM) [88] is a most widely used method to extract global features for age estimation and to provide information about the shape and appearance of a face [40], [79].…”

“…In 2016 Bukar et al [79] proposed supervised Appearance Model (sAM) for age and gender estimation that improves the AAM by using the partial least-squares (PLS) as the core of the model rather than PCA. They claimed that their proposed model (sAM) effectively represents the face features than AAM, whereas the PLS preserves worthy parts of the data that represent discriminatory features.…”

Gender Privacy: An Ensemble of Semi Adversarial Networks for Confounding Arbitrary Gender Classifiers

“…Extracting soft biometric attributes, such as age and gender from face images, has been extensively studied [9,16,13]. A wide range of methods have been employed, including those based on custom feature extraction techniques [10] and those based on deep learning techniques [26,29,20,12,16]. However, imparting soft biometric privacy by confounding such attributes is a relatively recent research area.…”

Gender and Age Classification of Human Faces for Automatic Detection of Anomalous Human Behaviour

“…Literature shows that there are two feature extraction methods have been used: local and global. The local approach is based on the part of the face such as wrinkles while global feature extraction focuses on the whole human face [12]. Feature points and geometry features are examples of local features.…”

“…PCA aims to reduce the dimension of the image [16]. AAM is the most widely used technique which uses statistical feature extraction method that takes both shapes and textures of the face images [12] into account. Guo & Huang [17] applied BIF to estimate human age from face images and achieved promising results.…”