Age group classification via structured fusion of uncertainty-driven shape features and selected surface features

Abstract

In this paper, we present a structured fusion method for facial age group classification. To utilize the structured fusion of shape features and surface features, we introduced the region of certainty (ROC) to not only control the classification accuracy for shape feature based system but also reduce the classification needs on surface feature based system. In the first stage, we design two shape features, which can be used to classify frontal faces with high accuracies. In the second stage, a surface feature is adopted and then selected by a statistical method. The statistical selected surface features combined with a SVM classifier can offer high classification rates. With properly adjusting the ROC by a single non-sensitive parameter, the structured fusion of two stages can provide a performance improvement. In the experiments, we use face images in the public available FG-NET and MORPH databases and partition them into three pre-defined age groups. It is observed that the proposed method offers a correct classification rate of 95.1% in FG-NET and 93.7% in MORPH, which outperforms state-ofthe-art methods by a significant margin.

Age estimation via face images: a survey

“…Using probabilistic neural network (PNN) to classify HOG features extracted from several regions, they achieved 87% accuracy in classifying face images into four groups. Liu et al [184] build a region of certainty (ROC) to link uncertainty-driven shape features with particular surface features. Two shape features are first designed to determine face certainty and classify it.…”

Hybridized Feature Extraction for Facial Age Estimation using SVM

“…Using probabilistic neural network (PNN) to classify HOG functions derived from multiple areas, 87 percent accuracy was achieved in the classification of face images into four groups. Liu et al [18] supplied age group classification through structured fusion of uncertainty-driven shape features and selected surface features. It was additionally discovered that the general performance of age-estimation diminishes as the quantity of age-group increases.…”

Age Classification from Facial Images: Is Frontalization Necessary?

“…With regard to image resolution, there is no preferred standard size. In many works [8,20,21], an image resolution of 60 × 60 has been used, in Liu et al [13] the image resolution is 180 × 150 and in Levi and Hassner [11] the image resolution is 256 × 256.…”

“…Levi and Hassner [11] also constrained their work to in-plane aligned images when feeding their convolutional neural network (CNN). Liu et al [13] described two new geometrical features (CirFace and Angle) which are Gaussianly distributed along certain age ranges and are defined on frontal images. Nevertheless, current databases, such as the FGnet Aging Database [14], contain images in the wild with variations in head pose.…”