Development of Bayesian Optimization Convolutional Neural Network model for Face Age Estimation

Abstract

Age estimation of unrestricted imaging circumstances has attracted an augmented recognition as it is applicable in several real-world applications such as security control, multimedia communication, human computer interaction, and surveillance. Age estimation is a process of determining the exact age or age group of a person depending on his biometric features. It is a challenging problem to effectively and automatically estimate ages of human. Recent research demonstrates that the deeply learned features for age estimation from large-scale data result in significant improvement of the age estimation performance for facial images. This research proposes a Convolutional Neural Network (CNN) using Bayesian Optimization for facial age estimation. Bayesian Optimization is applied to minimize the classification error on the validation set for CNN model. Also an enhanced model based on Gender Classification has been proposed as an extension to the previous model. As it is known, Males and Females have a variable type of appearance aging pattern; this results in age differently. This fact leads to assuming that using gender information may improve the age estimator performance for the preceding model. A Convolutional Neural Network (CNN) is used to get Gender Information, then Bayesian Optimization is applied to this pre-trained CNN when fine-tuned for age estimation task. Bayesian Optimization reduces the classification error on the validation set for the pre-trained model. Extensive experiments are done to evaluate the proposed model on three datasets: MORPH, FG-NET and FERET. The results show that using Bayesian Optimization on CNN outperforms the state-of-the-arts on FG-NET and FERET datasets with a Mean Absolute Error (MAE) of 2.88 and 1.3, and achieves good results compared to most of the state-of-the-art methods on MORPH dataset with a 3.16 MAE. Also, Extensive experiments are done to assess the enhanced model on two data sets: FERET and FG-NET. The experiments' result indicates that using a pre-trained CNN II containing Gender Information with Bayesian Optimization outperforms the state-of-the-arts on FERET and FG-NET data sets with a MAE of 1.2 and 2.67 respectively.