Joint Deep Learning of Facial Expression Synthesis and Recognition

被引:25
|
作者
Yan, Yan [1 ]
Huang, Ying [1 ]
Chen, Si [2 ]
Shen, Chunhua [3 ]
Wang, Hanzi [1 ]
机构
[1] Xiamen Univ, Fujian Key Lab Sensing & Comp Smart City, Sch Informat, Xiamen 361005, Peoples R China
[2] Xiamen Univ Technol, Sch Comp & Informat Engn, Xiamen 361024, Peoples R China
[3] Univ Adelaide, Sch Comp Sci, Adelaide, SA 5005, Australia
基金
中国国家自然科学基金; 国家重点研发计划;
关键词
Gallium nitride; Face recognition; Databases; Generative adversarial networks; Deep learning; Training data; Generators; Facial expression recognition; facial expression synthesis; convolutional neural networks (CNNs); generative adversarial net (GAN); NETWORKS; MANIFOLD; IMAGES;
D O I
10.1109/TMM.2019.2962317
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Recently, deep learning based facial expression recognition (FER) methods have attracted considerable attention and they usually require large-scale labelled training data. Nonetheless, the publicly available facial expression databases typically contain a small amount of labelled data. In this paper, to overcome the above issue, we propose a novel joint deep learning of facial expression synthesis and recognition method for effective FER. More specifically, the proposed method involves a two-stage learning procedure. Firstly, a facial expression synthesis generative adversarial network (FESGAN) is pre-trained to generate facial images with different facial expressions. To increase the diversity of the training images, FESGAN is elaborately designed to generate images with new identities from a prior distribution. Secondly, an expression recognition network is jointly learned with the pre-trained FESGAN in a unified framework. In particular, the classification loss computed from the recognition network is used to simultaneously optimize the performance of both the recognition network and the generator of FESGAN. Moreover, in order to alleviate the problem of data bias between the real images and the synthetic images, we propose an intra-class loss with a novel real data-guided back-propagation (RDBP) algorithm to reduce the intra-class variations of images from the same class, which can significantly improve the final performance. Extensive experimental results on public facial expression databases demonstrate the superiority of the proposed method compared with several state-of-the-art FER methods.
引用
收藏
页码:2792 / 2807
页数:16
相关论文
共 50 条
  • [21] Facial Depression Recognition by Deep Joint Label Distribution and Metric Learning
    Zhou, Xiuzhuang
    Wei, Zeqiang
    Xu, Min
    Qu, Shan
    Guo, Guodong
    IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, 2022, 13 (03) : 1605 - 1618
  • [22] Facial Expression Recognition via Joint Deep Learning of RGB-Depth Map Latent Representations
    Oyedotun, Oyebade K.
    Demisse, Girum
    Shabayek, Abd El Rahman
    Aouada, Djamila
    Ottersten, Bjorn
    2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW 2017), 2017, : 3161 - 3168
  • [23] Automated Facial Expression Recognition Framework Using Deep Learning
    Saeed, Saad
    Shah, Asghar Ali
    Ehsan, Muhammad Khurram
    Amirzada, Muhammad Rizwan
    Mahmood, Asad
    Mezgebo, Teweldebrhan
    JOURNAL OF HEALTHCARE ENGINEERING, 2022, 2022
  • [24] Deep Neural Networks with Relativity Learning for Facial Expression Recognition
    Guo, Yanan
    Tao, Dapeng
    Yu, Jun
    Xiong, Hao
    Li, Yaotang
    Tao, Dacheng
    2016 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA & EXPO WORKSHOPS (ICMEW), 2016,
  • [25] Automated Facial Expression Recognition Framework Using Deep Learning
    Saeed, Saad
    Shah, Asghar Ali
    Ehsan, Muhammad Khurram
    Amirzada, Muhammad Rizwan
    Mahmood, Asad
    Mezgebo, Teweldebrhan
    Journal of Healthcare Engineering, 2022, 2022
  • [26] Local Learning With Deep and Handcrafted Features for Facial Expression Recognition
    Georgescu, Mariana-Iuliana
    Ionescu, Radu Tudor
    Popescu, Marius
    IEEE ACCESS, 2019, 7 : 64827 - 64836
  • [27] Facial Expression Recognition System for Stress Detection with Deep Learning
    Almeida, Jose
    Rodrigues, Fatima
    PROCEEDINGS OF THE 23RD INTERNATIONAL CONFERENCE ON ENTERPRISE INFORMATION SYSTEMS (ICEIS 2021), VOL 1, 2021, : 256 - 263
  • [28] Facial expression recognition using lightweight deep learning modeling
    Ahmad, Mubashir
    Saira
    Alfandi, Omar
    Khattak, Asad Masood
    Qadri, Syed Furqan
    Saeed, Iftikhar Ahmed
    Khan, Salabat
    Hayat, Bashir
    Ahmad, Arshad
    MATHEMATICAL BIOSCIENCES AND ENGINEERING, 2023, 20 (05) : 8208 - 8225
  • [29] Facial Expression Recognition using Visual Saliency and Deep Learning
    Mavani, Viraj
    Raman, Shanmuganathan
    Miyapuram, Krishna P.
    2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW 2017), 2017, : 2783 - 2788
  • [30] Enhancing masked facial expression recognition with multimodal deep learning
    Shahzad, H. M.
    Bhatti, Sohail Masood
    Jaffar, Arfan
    Akram, Sheeraz
    MULTIMEDIA TOOLS AND APPLICATIONS, 2024, 83 (30) : 73911 - 73921