文献类型：期刊文献

英文题名：3D convolutional neural networks for facial expression classification

作者：Sun, Wenyun[1]; Zhao, Haitao[2]; Jin, Zhong[1]

机构：[1] School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China; [2] School of Information Science and Engineering, East China University of Science and Technology, Shanghai, China

年份：2017

卷号：10116 LNCS

起止页码：528

外文期刊名：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

收录：EI(收录号：20171303505478)

语种：英文

外文关键词：Convolution

摘要：In this paper, the general rules of designing 3D Convolutional Neural Networks are discussed. Four specific networks are designed for facial expression classification problem. Decisions of the four networks are fused together. The single networks and the ensemble network are evaluated on the extended Cohn-Kanade dataset, achieve accuracies of 92.31% and 96.15%. The performance outperform the state-of-the-art. A reusable open source project called 4DCNN is released. Based on this project, implementing 3D Convolutional Neural Networks for specific problems will be convenient. ? Springer International Publishing AG 2017.