Micro-expression recognition based on deep capsule adversarial domain adaptation network
Micro-expression recognition based on deep capsule adversarial domain adaptation network
复制标题
基于深度胶囊对抗域适应网络的微表情识别
DOI：
10.1117/1.jei.31.1.013021
复制
发表时间：
2022-01
影响因子：
1.1
通讯作者：
Hualin Zhan
中科院分区：
文献类型：
--
作者：
Zhihua Xie;Ling Shi;Sijia Cheng;Jiawei Fan;Hualin Zhan
Abstract. Micro-expression (ME), which reveals the genuine feelings and motives within human beings, attracts considerable attention in the field of automatic affective recognition. The main challenges for robust micro-expression recognition (MER) are from the short ME duration, low intensity of facial muscle movements, and insufficient samples. To meet these challenges, we propose an optical flow-based deep capsule adversarial domain adaptation network (DCADAN) for MER, which leverages a deep neural network stemming from these speculations. To alleviate the negative impact of the identity related features, optical flow preprocessing is applied to encode the subtle face motion information that is highly related to facial MEs. Then, a deep capsule network is developed to determine the part–whole relationships on optical flow features. To cope with the data deficiency and enhance the generalization capability via domain adaptation, an adversarial discriminator module that enriches the available samples from macro-expression data is integrated into the capsule network to train an expeditious end-to-end deep network. Finally, a simple and yet efficient attention module is embedded to the DCADAN to adaptively aggregate optical flow convolution maps into the primary capsule layers. We evaluate the performance of the entire network on the cross-database ME benchmark (3DB) using the leave-one-subject-out cross-validation. Unweighted F1-score (UF1) and unweighted average recall (UAR) are exploited as the evaluation metrics. The MER based on DCADAN achieves a UF1 score of 0.801 and a UAR score of 0.829 in comparison with a UF1 of 0.788 and a UAR of 0.782 for the updated approach. The comprehensive experimental results show that the incorporation of adversarial domain adaption into the capsule network is feasible and effective for representing discriminative features in ME and the proposed model outperforms state-of-the-art deep learning networks for MER.
登录
查看更多内容
跨数据库微表情识别：一个基准
影响因子：
8.9
作者：
Tong Zhang;Yuan Zong;Wenming Zheng;C. L. Philip Chen;Xiaopeng Hong;Chuangao Tang;Zhen Cui;Guoying Zhao
通讯作者：
Guoying Zhao
DOI：
10.1002/047134608x.w5513.pub2
发表时间：
2019-02
期刊：
Wiley Encyclopedia of Electrical and Electronics Engineering
影响因子：
--
作者：
K. Kulkarni;P. Turaga;Anuj Srivastava;Rama Chellappa
通讯作者：
K. Kulkarni;P. Turaga;Anuj Srivastava;Rama Chellappa
影响因子：
3.9
作者：
Lorenzo Vaquero;V. Brea;M. Mucientes
通讯作者：
Lorenzo Vaquero;V. Brea;M. Mucientes
DOI：
10.1049/iet-bmt.2018.5117
发表时间：
2016
期刊：
IET Biom.
影响因子：
--
作者：
Sheng He;Lambert Schomaker
通讯作者：
Sheng He;Lambert Schomaker
DOI：
10.1007/978-3-319-92198-3
发表时间：
2018
期刊：
--
影响因子：
--
作者：
Chao Sui;Bennamoun;R. Togneri
通讯作者：
Chao Sui;Bennamoun;R. Togneri