[go: up one dir, main page]
More Web Proxy on the site http://driver.im/ Skip to main content
Springer Nature Link
Log in

Compound facial expressions image generation for complex emotions

  • 1226: Deep-Patterns Emotion Recognition in the Wild
  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript

Abstract

This work presents the methodology to synthesize the complex facial expressions images from the learned representation without specifying emotion labels as input. The proposed methodology consists of three main modules: the basic emotion recognition model, linear regression, and the generative model. The recognition model is designed to extract the expression-related features that are the baseline for generation of complex facial expression. The linear regression is responsible for transforming expression features into latent space, which are taken by a generative model for image generation. In this work, two benchmark facial expressions datasets (Extended Cohn-Kanade and Japanese Female Facial Expressions) are used for the experiment. Based on our results, the proposed methodology provides the complex facial expressions images for compound emotions with comparatively high-visual quality. For quantitative assessment, the basic emotion recognition model can predict the an emotion from the generated compound facial expressions image by the proposed methodology with the accuracy of 67.51% and 62.87% respectively.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
£29.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price includes VAT (United Kingdom)

Instant access to the full article PDF.

Institutional subscriptions

Algorithm 1
Fig. 1
Fig. 2
Fig. 3
Algorithm 2
Fig. 4
Fig. 5

Similar content being viewed by others

Data Availability

The data generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

References

  1. Arjovsky M, Chintala S, Bottou L (2017) Wasserstein generative adversarial networks. In: International conference on machine learning, PMLR, pp 214–223

  2. Du S, Martinez AM (2015) Compound facial expressions of emotion: from basic research to clinical applications. Dialogues Clin Neurosci 17(4):443

    Article  Google Scholar 

  3. Du S, Tao Y, Martinez AM (2014) Compound facial expressions of emotion. Proc Natl Acad Sci 111(15):E1454–E1462

    Article  Google Scholar 

  4. Eckman P (1972) Universal and cultural differences in facial expression of emotion. In: Nebraska symposium on motivation, vol. 19, University of Nebraska Press, pp 207–284

  5. Ekman P, Friesen WV (1971) Constants across cultures in the face and emotion. J Pers Soc Psychol 17(2):124

    Article  Google Scholar 

  6. Ekman P, Oster H (1979) Facial expressions of emotion. Annu Rev Psychol 30(1):527–554

    Article  Google Scholar 

  7. Friesen E, Ekman P (1978) Facial action coding system: a technique for the measurement of facial movement. Palo Alto 3(2):5

    Google Scholar 

  8. Geitgey A (2019) Face recognition documentation. Release 1. 2 (3):3–37

    Google Scholar 

  9. Ghatas FS, Hemayed EE (2020) GANKIN: Generating Kin faces using disentangled GAN. SN Appl Sci 2:166. https://doi.org/10.1007/s42452-020-1949-3

    Article  Google Scholar 

  10. Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Bengio Y (2014) Generative adversarial nets. Adv Neural Inf Process Syst :27

  11. Gulrajani I, Ahmed F, Arjovsky M, Dumoulin V, Courville A (2017) Improved training of wasserstein gans. arXiv:1704.00028

  12. Heusel M, Ramsauer H, Unterthiner T, Nessler B, Hochreiter S (2017) Gans trained by a two time-scale update rule converge to a local nash equilibrium. Adv Neural Inform Process Syst 30:

  13. Hupka RB (1984) Jealousy: compound emotion or label for a particular situation? Motiv Emot 8(2):141–155. https://doi.org/10.1007/BF00993070

    Article  Google Scholar 

  14. Karras T, Aila T, Laine S, Lehtinen J (2017) Progressive growing of gans for improved quality, stability, and variation. arXiv:1710.10196

  15. Karras T, Miika A, Janne H, Samuli L, Jaakko L, Timo A (2020) Training generative adversarial networks with limited data. Adv Neural Inf Process Syst 33:12104–12114

    Google Scholar 

  16. Kazemi V, Sullivan J (2014) One millisecond face alignment with an ensemble of regression trees. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1867–1874, https://doi.org/10.1109/CVPR.2014.241

  17. Khine WSS, Siritanawan P, Kotani K (2020) Generation of compound emotions expressions with emotion generative adversarial networks (EmoGANs). In: 2020 59th Annual conference of the society of instrument and control engineers of Japan (SICE), IEEE, pp 748–755, https://doi.org/10.23919/SICE48898.2020.9240306

  18. Khine WSS, Siritanawan P, Kotani K (2021) Wasserstein based EmoGANs+. In: 2021 Joint 10th international conference on informatics, electronics and vision (ICIEV) and 2021 5th International conference on imaging, vision and pattern recognition (icIVPR), IEEE, pp 1–8, https://doi.org/10.1109/ICIEVicIVPR52578.2021.9564216

  19. King DE (2009) Dlib-ml: a machine learning toolkit. J Mach Learn Res 10:1755–1758. https://doi.org/10.5555/1577069.1755843

    Article  Google Scholar 

  20. Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. arXiv:1412.6980

  21. LaPlante D, Ambady N (2000) Multiple messages: facial recognition advantage for compound expressions. J Nonverbal Behav 24(3):211–224. https://doi.org/10.1023/A:1006641104653

    Article  Google Scholar 

  22. LinderNorén E (2021) Keras-GAN. https://github.com/eriklindernoren/Keras-GAN

  23. Lucey P, Cohn JF, Kanade T, Saragih J, Ambadar Z, Matthews I (2010) The Extended Cohn-Kanade Dataset (CK+): A complete dataset for action unit and emotion-specified expression. In: 2010 IEEE computer society conference on computer vision and pattern recognition - workshops, pp 94–101, https://doi.org/10.1109/CVPRW.2010.5543262

  24. Lyons M, Kamachi M, Gyoba J (2017) Japanese female facial expression (JAFFE) database

  25. Martinez A, Du S (2012) A model of the perception of facial expressions of emotion by humans: research overview and perspectives. J Mach Learn Res 13(5):

  26. Mehrabian A, Ferris SR (1967) Inference of attitudes from nonverbal communication in two channels. J Consult Psychol 31(3):248

    Article  Google Scholar 

  27. Mirza M, Osindero S (2014) Conditional generative adversarial nets. arXiv:1411.1784

  28. Odena A (2016) Semi-supervised learning with generative adversarial networks. arXiv:1606.01583

  29. Parent A (2005) Duchenne de boulogne: a pioneer in neurology and medical photography. Can J Neurol Sci 32(3):369–377. https://doi.org/10.1017/S031716710000431

    Article  Google Scholar 

  30. Radford A, Metz L, Chintala S (2015) Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv:1511.06434

  31. Tan M, Le Q (2019) Efficientnet: rethinking model scaling for convolutional neural networks. In: International conference on machine learning, PMLR, pp 6105–6114

Download references

Funding

This work was supported by Japan Advanced Institute of Science and Technology Research Grants (Houga).

Author information

Author notes

    Authors and Affiliations

    1. School of Information Science, Japan Advanced Institute of Science and Technology, Nomi, 9231211, Ishikawa, Japan

      Shwe Sin Khine Win, Prarinya Siritanawan & Kazunori Kotani

    2. Division of Transdisciplinary Science, Japan Advanced Institute of Science and Technology, Nomi, 9231211, Ishikawa, Japan

      Kazunori Kotani

    Authors
    1. Shwe Sin Khine Win
      View author publications

      You can also search for this author in PubMed Google Scholar

    2. Prarinya Siritanawan
      View author publications

      You can also search for this author in PubMed Google Scholar

    3. Kazunori Kotani
      View author publications

      You can also search for this author in PubMed Google Scholar

    Corresponding authors

    Correspondence to Shwe Sin Khine Win, Prarinya Siritanawan or Kazunori Kotani.

    Ethics declarations

    Competing interests

    The authors have no competing interests to declare that are relevant to the content of this article.

    Additional information

    Publisher’s note

    Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

    Shwe Sin Khine Win, Prarinya Siritanawan and Kazunori Kotani are contributed equally to this work.

    Appendix: Additional results

    Appendix: Additional results

    Fig. 6
    figure 6

    Basic facial expression recognition model training in CK+ dataset [23]

    Full size image
    Fig. 7
    figure 7

    Basic facial expression recognition model training in JAFFE dataset [24]

    Full size image
    Fig. 8
    figure 8

    Confusion matrix over test set

    Full size image
    Table 8 Examples of Generated Images By GANs, DCGANs, cGANs
    Full size table
    Table 9 Generated images with CK+ feature extractor, f (.) and retrained generator, G(.) with CK+ dataset [23]
    Full size table
    Table 10 Generated images with JAFFE feature extractor, f (.) and retrained generator, G(.) with JAFFE dataset [24]
    Full size table

    Rights and permissions

    Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

    Reprints and permissions

    About this article

    Check for updates. Verify currency and authenticity via CrossMark

    Cite this article

    Win, S.S.K., Siritanawan, P. & Kotani, K. Compound facial expressions image generation for complex emotions. Multimed Tools Appl 82, 11549–11588 (2023). https://doi.org/10.1007/s11042-022-14289-7

    Download citation

    • Received:

    • Revised:

    • Accepted:

    • Published:

    • Issue Date:

    • DOI: https://doi.org/10.1007/s11042-022-14289-7

    Keywords

    Search

    Navigation