Spatial-Temporal Recurrent Neural Network for Emotion Recognition

被引:258
|
作者
Zhang, Tong [1 ,2 ]
Zheng, Wenming [3 ]
Cui, Zhen [4 ]
Zong, Yuan [3 ]
Li, Yang [1 ,2 ]
机构
[1] Southeast Univ, Key Lab Child Dev & Learning Sci, Minist Educ, Nanjing 210096, Jiangsu, Peoples R China
[2] Southeast Univ, Dept Informat Sci & Engn, Nanjing 210096, Jiangsu, Peoples R China
[3] Southeast Univ, Res Ctr Learning Sci, Minist Educ, Key Lab Child Dev & Learning Sci, Nanjing 210096, Jiangsu, Peoples R China
[4] Nanjing Univ Sci & Technol, Sch Comp Sci & Engn, Nanjing 210094, Jiangsu, Peoples R China
基金
中国国家自然科学基金;
关键词
Electroencephalogram (EEG) emotion recognition; emotion recognition; facial expression recognition; spatial- temporal recurrent neural network (STRNN);
D O I
10.1109/TCYB.2017.2788081
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In this paper, we propose a novel deep learning framework, called spatial-temporal recurrent neural network (STRNN), to integrate the feature learning from both spatial and temporal information of signal sources into a unified spatial-temporal dependency model. In STRNN, to capture those spatially co-occurrent variations of human emotions, a multidirectional recurrent neural network (RNN) layer is employed to capture long-range contextual cues by traversing the spatial regions of each temporal slice along different directions. Then a hi-directional temporal RNN layer is further used to learn the discriminative features characterizing the temporal dependencies of the sequences, where sequences are produced from the spatial RNN layer. To further select those salient regions with more discriminative ability for emotion recognition, we impose sparse projection onto those hidden states of spatial and temporal domains to improve the model discriminant ability. Consequently, the proposed two-layer RNN model provides an effective way to make use of both spatial and temporal dependencies of the input signals for emotion recognition. Experimental results on the public emotion datasets of electroencephalogram and facial expression demonstrate the proposed STRNN method is more competitive over those state-of-the-art methods.
引用
收藏
页码:839 / 847
页数:9
相关论文
共 50 条
  • [41] EEG-based emotion recognition using a temporal-difference minimizing neural network
    Xiangyu Ju
    Ming Li
    Wenli Tian
    Dewen Hu
    Cognitive Neurodynamics, 2024, 18 : 405 - 416
  • [42] GC-STCL: A Granger Causality-Based Spatial-Temporal Contrastive Learning Framework for EEG Emotion Recognition
    Wang, Lei
    Wang, Siming
    Jin, Bo
    Wei, Xiaopeng
    ENTROPY, 2024, 26 (07)
  • [43] A deep neural network model of the primate superior colliculus for emotion recognition
    Mendez, Carlos Andres
    Celeghin, Alessia
    Diano, Matteo
    Orsenigo, Davide
    Ocak, Brian
    Tamietto, Marco
    PHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCIETY B-BIOLOGICAL SCIENCES, 2022, 377 (1863)
  • [44] Emotion Recognition by Facial Features using Recurrent Neural Networks
    Mostafa, Amr
    Khalil, Mahmoud I.
    Abbas, Hazem
    PROCEEDINGS OF 2018 13TH INTERNATIONAL CONFERENCE ON COMPUTER ENGINEERING AND SYSTEMS (ICCES), 2018, : 417 - 422
  • [45] Semantic-Emotion Neural Network for Emotion Recognition From Text
    Batbaatar, Erdenebileg
    Li, Meijing
    Ryu, Keun Ho
    IEEE ACCESS, 2019, 7 : 111866 - 111878
  • [46] Emotion Recognition in Reddit Comments Using Recurrent Neural Networks
    Rezapour M.
    Recent Advances in Computer Science and Communications, 2024, 17 (04) : 79 - 87
  • [47] Emotion recognition by skeleton-based spatial and temporal analysis
    Oguz, Abdulhalik
    Ertugrul, Omer Faruk
    EXPERT SYSTEMS WITH APPLICATIONS, 2024, 238
  • [48] LONG SHORT TERM MEMORY RECURRENT NEURAL NETWORK BASED ENCODING METHOD FOR EMOTION RECOGNITION IN VIDEO
    Chao, Linlin
    Tao, Jianhua
    Yang, Minghao
    Li, Ya
    Wen, Zhengqi
    2016 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING PROCEEDINGS, 2016, : 2752 - 2756
  • [49] Emotion Recognition from Multi-Channel EEG Data through Convolutional Recurrent Neural Network
    Li, Xiang
    Song, Dawei
    Zhang, Peng
    Yu, Guangliang
    Hou, Yuexian
    Hu, Bin
    2016 IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE (BIBM), 2016, : 352 - 359
  • [50] EEG-based emotion recognition using 4D convolutional recurrent neural network
    Fangyao Shen
    Guojun Dai
    Guang Lin
    Jianhai Zhang
    Wanzeng Kong
    Hong Zeng
    Cognitive Neurodynamics, 2020, 14 : 815 - 828