Skip to main content
Log in

Font generation based on least squares conditional generative adversarial nets

  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript


With the rapid growth of multimedia information, the font library has become a part of people’s work life. Compared to the Western alphabet language, it is difficult to create new font due to huge quantity and complex shape. At present, most of the researches on automatic generation of fonts use traditional methods requiring a large number of rules and parameters set by experts, which are not widely adopted. This paper divides Chinese characters into strokes and generates new font strokes by fusing the styles of two existing font strokes and assembling them into new fonts. This approach can effectively improve the efficiency of font generation, reduce the costs of designers, and is able to inherit the style of existing fonts. In the process of learning to generate new fonts, the popular of deep learning areas, Generative Adversarial Nets has been used. Compared with the traditional method, it can generate higher quality fonts without well-designed and complex loss function.

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

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others




  1. Arjovsky M, Chintala S, Bottou L (2017) Wasserstein GAN. arXiv:1701.07875v2

  2. Chen X, Duan Y, Houthooft R et al (2016) InfoGAN: interpretable representation learning by information maximizing generative adversarial nets. Neural Information Processing Systems (NIPS)

  3. Chuang Y-N, Huang Z-Y, Tsai Y-L (2017) Variational grid setting network. arXiv preprint arXiv:1710.01255

  4. Denton E, Chintala S, Szlam A et al (2015) Deep generative image models using a Laplacian pyramid of adversarial networks. arXiv:1506.05751

  5. Gatys LA, Ecker AS, Bethge M (2015) A neural algorithm of artistic style. arXiv:1508.06576v2

  6. Goodfellow IJ, Pougetabadie J, Mirza M et al (2014) Generative Adversarial Nets. Adv Neural Inf Proces Syst 3:2672–2680

    Google Scholar 

  7. Gregor K, Danihelka I, Graves A et al. (2015) DRAW: a recurrent neural network for image generation. arXiv:1502.04623v2

  8. He K, Zhang X, Ren S et al (2016) Deep residual learning for image recognition. 2016 IEEE conference on computer vision and pattern recognition (CVPR), pp 770-778

  9. Isola P, Zhu JY, Zhou T et al (2016) Image-to-image translation with conditional adversarial networks. arXiv:1611.07004v1

  10. Johnson J, Alahi A, Li FF (2016) Perceptual losses for real-time style transfer and super-resolution. arXiv:1603.08155v1

  11. Krizhevsky A, Sutskever I, Hinton GE (2012) ImageNet classification with deep convolutional neural networks. International conference on neural information processing systems. Curran associates Inc., pp 1097-1105

  12. Li C, Jiangqing W, Bo L et al (2013) Algorithm on strokes separation for Chinese characters based on edge. Computer. Science 40(7):307–311

    Google Scholar 

  13. Lian Z, Zhao B, Xiao J (2016) Automatic generation of large-scale handwriting fonts via style learning[C]// SIGGRAPH ASIA 2016 technical briefs. ACM, pp 12

  14. Liu MY, Tuzel O (2016) Coupled generative adversarial networks. arXiv:1606.07536v2

  15. Lyu P et al (2017) Auto-encoder guided GAN for Chinese calligraphy synthesis. arXiv preprint arXiv:1706.08789

  16. Mao X, Li Q, Xie H et al (2017) Least squares generative adversarial networks. arXiv:1611.04076v3

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

  18. Mouchère H et al (2013) A dynamic time warping algorithm for recognition of multi-stroke on-line Handwriten characters. Journal of South China University of Technology 41(7).

  19. Myronenko A, Song X (2010) Point set registration: coherent point drift. IEEE Trans Pattern Anal Mach Intell 32(12):2262–2275

    Article  Google Scholar 

  20. Odena A (2016) Semi-supervised learning with generative adversarial networks. International conference on machine learning (ICML)

  21. Odena A, Olah C, Shlens J (2016) Conditional image synthesis with auxiliary classifier GANs. arXiv:1610.09585

  22. Radford A, Metz L, Chintala S (2015) Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv:1511.06434v2

  23. Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556

  24. Springenberg JT (2015) Unsupervised and semi-supervised learning with categorical generative adversarial networks. arXiv:1511.06390v2

  25. Szegedy C, Liu W, Jia Y et al (2015) Going deeper with convolutions. 2015 IEEE conference on computer vision and pattern recognition (CVPR), pp 1-9

  26. Taigman Y, Polyak A, Wolf L (2016) Unsupervised cross-domain image generation. arXiv:1611.02200v1

  27. Tan CL, Cao R (2000) A model of stroke extraction from Chinese character images. In: Proceedings 15th International Conference on Pattern Recognition, Barcelona, Spain, vol 4, pp 368–371

  28. Wu L, Xia Y, Zhao L et al (2017) Adversarial neural machine translation. arXiv:1704.06933v3

  29. Xiafe Z, Jiayan L et al (2016) Extracting Chinese calligraphy strokes using stroke crawler. Journal of Computer-Aided Design & Computer Graphics 28(2):301–309

    Google Scholar 

  30. Xiaohu M, Yulong L, Zhigeng P et al (1999) The automatic generation for Chinese outline font and It’s transformation method. Journal of Chinese Information Processing 13(2):46–50

    Google Scholar 

  31. Xu S, Jin T, Jiang H et al (2009) Automatic Generation of Personal Chinese Handwriting by Capturing the Characteristics of Personal Handwriting. Conference on Innovative Applications of Artificial Intelligence, July 14–16, 2009, Pasadena, California, USA. DBLP

  32. Zhang XY, Yin F, Zhang YM et al (2017) Drawing and recognizing Chinese characters with recurrent neural network. TPAMI

  33. Zhao S et al (1949) Predicting personalized image emotion perceptions in social networks. IEEE Trans Affect Comput 99:1–1

    Google Scholar 

  34. Zhao S, Yao H, Sun X (2011) Affective video classification based on Spatio-temporal feature fusion. Sixth international conference on image and graphics IEEE computer. Society:795–800

  35. Zhao S et al (2015) Strategy for dynamic 3D depth data matching towards robust action retrieval. Neurocomputing 151:533–543

    Article  Google Scholar 

  36. Zhao S et al (2016) Predicting personalized emotion perceptions of social images. ACM on multimedia conference ACM, pp 1385-1394

  37. Zhao S et al (2017) Continuous probability distribution prediction of image emotions via multi-task shared sparse regression. IEEE Trans Multimedia 99:1–1

    Article  Google Scholar 

  38. Zong A, Zhu Y (2014) StrokeBank: automating personalized chinese handwriting generation twenty-eighth AAAI conference on artificial intelligence. AAAI Press, Palo Alto, pp 3024–3029

    Google Scholar 

Download references


This work is supported by the National Key Technology R&D Program (No. 2017YFC011300, No. 2016YFB1001503), the Nature Science Foundation of China (No. 61422210, No. 61373076, No. 61402388, and No. 61572410),the Nature Science Foundation of Fujian Province, China (No. 2017 J01125).

Author information

Authors and Affiliations


Corresponding author

Correspondence to Rongrong Ji.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lin, X., Li, J., Zeng, H. et al. Font generation based on least squares conditional generative adversarial nets. Multimed Tools Appl 78, 783–797 (2019).

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: