Yang H.-FLin B.-YChang K.-YCHU-SONG CHEN2021-09-022021-09-02201815516857https://www.scopus.com/inward/record.uri?eid=2-s2.0-85042552102&doi=10.1145%2f3152118&partnerID=40&md5=c006c617f78e9ae63d37167049d54f05https://scholars.lib.ntu.edu.tw/handle/123456789/581327This article tackles the problem of joint estimation of human age and facial expression. This is an important yet challenging problem because expressions can alter face appearances in a similar manner to human aging. Different from previous approaches that deal with the two tasks independently, our approach trains a convolutional neural network (CNN) model that unifies ordinal regression and multi-class classification in a single framework. We demonstrate experimentally that our method performs more favorably against state-of-the-art approaches. ? 2018 ACM.Convolution; Learning systems; Neural networks; Age estimation; Convolutional networks; Expression recognition; Multilevels; Multitask learning; Scattering networks; Transfer learning; Deep learning[SDGs]SDG10journal article10.1145/31521182-s2.0-85042552102