Practical age estimation using deep label distribution learning

作     者：Huiying ZHANG Yu ZHANG Xin GENG Huiying ZHANG;Yu ZHANG;Xin GENG

作者机构：Pujiang InstituteNanjing Tech UniversityNanjing 211200China School of Computer Science and EngineeringSoutheast UniversityNanjing 211189China 

出 版 物：《Frontiers of Computer Science》 (中国计算机科学前沿（英文版）)

年 卷 期：2021年第15卷第3期

页      面：75-80页

核心收录：

学科分类：081203[工学-计算机应用技术] 08[工学] 0835[工学-软件工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：the financial support of the China National Natural Science Foundation(61702095) Natural Science Founda-tion(njpj2018209)of Nanjing Tech University Pujiang Institute,Anhui Polytechnic University Scientific Research Foundation(S031702004) Natural Science Foundation of Fujian Province(2018J01806) Scientific Research Pro-gram of Outstanding Talents in Universities of Fujian 

主　　题：deep learning convolutional neural networks label distribution learning facial age estimation 

摘      要：Age estimation plays an important role in human-computer interaction *** lack of large number of facial images with definite age label makes age estimation al-gorithms *** label distribution learning(DLDL)which employs convolutional neural networks(CNN)and label distribution learning to learn ambiguity from ground-truth age and adjacent ages,has been proven to outperform current state-of-the-art ***,DLDL assumes a rough label distribution which covers all ages for any given age *** this paper,a more practical label distribution paradigm is proposed:we limit age label distribution that only covers a reasonable number of neighboring *** addition,we explore different label distributions to improve the performance of the proposed learning *** employ CNN and the improved label distribution learning to estimate *** results show that compared to the DLDL,our method is more effective for facial age recognition.