Joint Multi-feature Learning for Facial Age Estimation

Age estimation from face images has attracted much attention due to its favorable of many real-world applications such as video surveillance and social networking. However, most existing studies usually directly extract aging-feature, which ignore the high age-related factors such as race and gender information. In this paper, we propose a joint multi-feature learning method for robust facial age estimation by extensively exploring age-related features. Specifically, we first specially learn the race and gender features from face images, which are two highly related information for age estimation of an individual. Then, we jointly learn the aging-feature by concatenating these race-specific and gender-specific information maps with the original face images. To fully utilize the continuity and the order of age labels, we form a regression-ranking age estimator to predict the final age. Experimental results on three benchmark databases demonstrate the superior performance of our proposed method on facial age estimation in comparison with other stateof-the-art methods.